Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[BugFix] Fix calls to _reset_env_preprocess #2798

Merged
merged 1 commit into from
Feb 20, 2025

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 20, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2798

Note: Links to docs will display an error until the docs builds have been completed.

❌ 11 New Failures, 1 Unrelated Failure

As of commit 90c51f7 with merge base 76aa9bc (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

This comment was automatically generated by Dr. CI and updates every 15 minutes.

vmoens added a commit that referenced this pull request Feb 20, 2025
ghstack-source-id: 59925635a87b196a5bcb0fb251afe4cc7b8b103e
Pull Request resolved: #2798
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 20, 2025
Copy link

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 149. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.6231s 0.5356s 1.8669 Ops/s 1.9662 Ops/s $\textbf{\color{#d91a1a}-5.05\%}$
test_transformed 1.1599s 1.0567s 0.9463 Ops/s 1.0308 Ops/s $\textbf{\color{#d91a1a}-8.20\%}$
test_serial 1.6356s 1.5561s 0.6426 Ops/s 0.6700 Ops/s $\color{#d91a1a}-4.09\%$
test_parallel 1.4056s 1.3151s 0.7604 Ops/s 0.7563 Ops/s $\color{#35bf28}+0.54\%$
test_step_mdp_speed[True-True-True-True-True] 0.1541ms 30.1482μs 33.1695 KOps/s 32.2834 KOps/s $\color{#35bf28}+2.74\%$
test_step_mdp_speed[True-True-True-True-False] 66.4250μs 17.8772μs 55.9373 KOps/s 56.4082 KOps/s $\color{#d91a1a}-0.83\%$
test_step_mdp_speed[True-True-True-False-True] 57.2770μs 17.1683μs 58.2469 KOps/s 58.1458 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-True-True-False-False] 57.8280μs 10.0195μs 99.8056 KOps/s 99.9509 KOps/s $\color{#d91a1a}-0.15\%$
test_step_mdp_speed[True-True-False-True-True] 68.7480μs 32.1155μs 31.1376 KOps/s 30.3870 KOps/s $\color{#35bf28}+2.47\%$
test_step_mdp_speed[True-True-False-True-False] 68.4970μs 19.6258μs 50.9533 KOps/s 51.0904 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[True-True-False-False-True] 47.4280μs 18.9980μs 52.6372 KOps/s 51.9147 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[True-True-False-False-False] 49.5620μs 11.8614μs 84.3071 KOps/s 83.2314 KOps/s $\color{#35bf28}+1.29\%$
test_step_mdp_speed[True-False-True-True-True] 90.5780μs 34.3654μs 29.0991 KOps/s 28.7332 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[True-False-True-True-False] 62.3260μs 21.6489μs 46.1918 KOps/s 46.5911 KOps/s $\color{#d91a1a}-0.86\%$
test_step_mdp_speed[True-False-True-False-True] 70.8110μs 19.0617μs 52.4612 KOps/s 51.8566 KOps/s $\color{#35bf28}+1.17\%$
test_step_mdp_speed[True-False-True-False-False] 54.4410μs 11.8180μs 84.6164 KOps/s 83.7201 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[True-False-False-True-True] 97.5710μs 36.2882μs 27.5571 KOps/s 26.3472 KOps/s $\color{#35bf28}+4.59\%$
test_step_mdp_speed[True-False-False-True-False] 0.3963ms 24.2713μs 41.2009 KOps/s 42.9616 KOps/s $\color{#d91a1a}-4.10\%$
test_step_mdp_speed[True-False-False-False-True] 0.2796ms 22.2406μs 44.9629 KOps/s 47.0171 KOps/s $\color{#d91a1a}-4.37\%$
test_step_mdp_speed[True-False-False-False-False] 51.5780μs 13.6587μs 73.2135 KOps/s 73.6760 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[False-True-True-True-True] 66.5840μs 34.1376μs 29.2932 KOps/s 28.7046 KOps/s $\color{#35bf28}+2.05\%$
test_step_mdp_speed[False-True-True-True-False] 72.7250μs 21.5698μs 46.3611 KOps/s 46.3637 KOps/s $-0.01\%$
test_step_mdp_speed[False-True-True-False-True] 65.6520μs 21.7178μs 46.0452 KOps/s 45.1893 KOps/s $\color{#35bf28}+1.89\%$
test_step_mdp_speed[False-True-True-False-False] 41.4570μs 13.3163μs 75.0957 KOps/s 73.8166 KOps/s $\color{#35bf28}+1.73\%$
test_step_mdp_speed[False-True-False-True-True] 87.4230μs 36.1037μs 27.6980 KOps/s 27.1551 KOps/s $\color{#35bf28}+2.00\%$
test_step_mdp_speed[False-True-False-True-False] 59.0290μs 23.1895μs 43.1229 KOps/s 42.7122 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[False-True-False-False-True] 2.6769ms 23.6545μs 42.2753 KOps/s 41.1965 KOps/s $\color{#35bf28}+2.62\%$
test_step_mdp_speed[False-True-False-False-False] 0.1651ms 15.0274μs 66.5451 KOps/s 66.1165 KOps/s $\color{#35bf28}+0.65\%$
test_step_mdp_speed[False-False-True-True-True] 0.1417ms 37.8275μs 26.4358 KOps/s 25.8216 KOps/s $\color{#35bf28}+2.38\%$
test_step_mdp_speed[False-False-True-True-False] 73.0960μs 25.4574μs 39.2814 KOps/s 40.0113 KOps/s $\color{#d91a1a}-1.82\%$
test_step_mdp_speed[False-False-True-False-True] 60.2320μs 23.7087μs 42.1787 KOps/s 40.6078 KOps/s $\color{#35bf28}+3.87\%$
test_step_mdp_speed[False-False-True-False-False] 49.7420μs 15.2363μs 65.6325 KOps/s 66.6175 KOps/s $\color{#d91a1a}-1.48\%$
test_step_mdp_speed[False-False-False-True-True] 83.2150μs 39.7919μs 25.1307 KOps/s 24.4343 KOps/s $\color{#35bf28}+2.85\%$
test_step_mdp_speed[False-False-False-True-False] 71.5830μs 26.7591μs 37.3705 KOps/s 36.0642 KOps/s $\color{#35bf28}+3.62\%$
test_step_mdp_speed[False-False-False-False-True] 97.4210μs 25.2691μs 39.5741 KOps/s 38.4211 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[False-False-False-False-False] 74.9400μs 16.6804μs 59.9504 KOps/s 59.2791 KOps/s $\color{#35bf28}+1.13\%$
test_values[generalized_advantage_estimate-True-True] 10.2023ms 9.6442ms 103.6890 Ops/s 102.4127 Ops/s $\color{#35bf28}+1.25\%$
test_values[vec_generalized_advantage_estimate-True-True] 27.2800ms 24.4610ms 40.8814 Ops/s 37.5441 Ops/s $\textbf{\color{#35bf28}+8.89\%}$
test_values[td0_return_estimate-False-False] 0.2519ms 0.1942ms 5.1488 KOps/s 4.9151 KOps/s $\color{#35bf28}+4.75\%$
test_values[td1_return_estimate-False-False] 27.0558ms 23.9854ms 41.6920 Ops/s 41.0469 Ops/s $\color{#35bf28}+1.57\%$
test_values[vec_td1_return_estimate-False-False] 26.6854ms 24.3793ms 41.0184 Ops/s 37.6713 Ops/s $\textbf{\color{#35bf28}+8.88\%}$
test_values[td_lambda_return_estimate-True-False] 36.9775ms 34.7367ms 28.7880 Ops/s 28.5640 Ops/s $\color{#35bf28}+0.78\%$
test_values[vec_td_lambda_return_estimate-True-False] 27.1563ms 24.3711ms 41.0323 Ops/s 37.8247 Ops/s $\textbf{\color{#35bf28}+8.48\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.5974ms 8.3931ms 119.1461 Ops/s 117.1228 Ops/s $\color{#35bf28}+1.73\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4435ms 1.9218ms 520.3529 Ops/s 513.8792 Ops/s $\color{#35bf28}+1.26\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6164ms 0.3734ms 2.6782 KOps/s 2.7664 KOps/s $\color{#d91a1a}-3.19\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 47.6268ms 44.1673ms 22.6412 Ops/s 22.4286 Ops/s $\color{#35bf28}+0.95\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.3737ms 3.4402ms 290.6796 Ops/s 287.7728 Ops/s $\color{#35bf28}+1.01\%$
test_dqn_speed[False-None] 2.0459ms 1.4411ms 693.9224 Ops/s 711.6841 Ops/s $\color{#d91a1a}-2.50\%$
test_dqn_speed[False-backward] 2.0421ms 1.9357ms 516.6076 Ops/s 528.9615 Ops/s $\color{#d91a1a}-2.34\%$
test_dqn_speed[True-None] 0.8322ms 0.4953ms 2.0190 KOps/s 1.9467 KOps/s $\color{#35bf28}+3.71\%$
test_dqn_speed[True-backward] 0.9946ms 0.9314ms 1.0737 KOps/s 1.0350 KOps/s $\color{#35bf28}+3.74\%$
test_dqn_speed[reduce-overhead-None] 0.7168ms 0.4974ms 2.0104 KOps/s 1.9707 KOps/s $\color{#35bf28}+2.02\%$
test_dqn_speed[reduce-overhead-backward] 0.9651ms 0.9339ms 1.0708 KOps/s 1.0032 KOps/s $\textbf{\color{#35bf28}+6.74\%}$
test_ddpg_speed[False-None] 3.7444ms 2.9686ms 336.8625 Ops/s 344.6263 Ops/s $\color{#d91a1a}-2.25\%$
test_ddpg_speed[False-backward] 4.2338ms 4.0823ms 244.9585 Ops/s 247.2634 Ops/s $\color{#d91a1a}-0.93\%$
test_ddpg_speed[True-None] 2.1044ms 1.2558ms 796.3131 Ops/s 800.1348 Ops/s $\color{#d91a1a}-0.48\%$
test_ddpg_speed[True-backward] 2.2363ms 2.1572ms 463.5604 Ops/s 435.3784 Ops/s $\textbf{\color{#35bf28}+6.47\%}$
test_ddpg_speed[reduce-overhead-None] 1.9026ms 1.2535ms 797.7642 Ops/s 783.1370 Ops/s $\color{#35bf28}+1.87\%$
test_ddpg_speed[reduce-overhead-backward] 2.2191ms 2.1599ms 462.9940 Ops/s 458.9103 Ops/s $\color{#35bf28}+0.89\%$
test_sac_speed[False-None] 8.7978ms 8.2002ms 121.9487 Ops/s 123.6630 Ops/s $\color{#d91a1a}-1.39\%$
test_sac_speed[False-backward] 12.1288ms 11.4519ms 87.3215 Ops/s 91.0137 Ops/s $\color{#d91a1a}-4.06\%$
test_sac_speed[True-None] 2.7852ms 2.1256ms 470.4453 Ops/s 461.0633 Ops/s $\color{#35bf28}+2.03\%$
test_sac_speed[True-backward] 3.9143ms 3.8080ms 262.6066 Ops/s 239.5265 Ops/s $\textbf{\color{#35bf28}+9.64\%}$
test_sac_speed[reduce-overhead-None] 2.6897ms 2.1559ms 463.8534 Ops/s 460.1507 Ops/s $\color{#35bf28}+0.80\%$
test_sac_speed[reduce-overhead-backward] 3.9062ms 3.8085ms 262.5723 Ops/s 248.5682 Ops/s $\textbf{\color{#35bf28}+5.63\%}$
test_redq_speed[False-None] 14.1443ms 13.0052ms 76.8926 Ops/s 71.2803 Ops/s $\textbf{\color{#35bf28}+7.87\%}$
test_redq_speed[False-backward] 25.6131ms 22.5117ms 44.4214 Ops/s 42.4926 Ops/s $\color{#35bf28}+4.54\%$
test_redq_speed[True-None] 5.4361ms 4.8174ms 207.5800 Ops/s 157.5772 Ops/s $\textbf{\color{#35bf28}+31.73\%}$
test_redq_speed[True-backward] 13.1727ms 12.0605ms 82.9154 Ops/s 71.1006 Ops/s $\textbf{\color{#35bf28}+16.62\%}$
test_redq_speed[reduce-overhead-None] 5.9814ms 4.8255ms 207.2317 Ops/s 160.5812 Ops/s $\textbf{\color{#35bf28}+29.05\%}$
test_redq_speed[reduce-overhead-backward] 13.0046ms 12.5690ms 79.5611 Ops/s 70.4657 Ops/s $\textbf{\color{#35bf28}+12.91\%}$
test_redq_deprec_speed[False-None] 23.6679ms 13.4479ms 74.3610 Ops/s 71.2793 Ops/s $\color{#35bf28}+4.32\%$
test_redq_deprec_speed[False-backward] 21.3084ms 18.6216ms 53.7010 Ops/s 48.4801 Ops/s $\textbf{\color{#35bf28}+10.77\%}$
test_redq_deprec_speed[True-None] 4.3288ms 3.8837ms 257.4840 Ops/s 218.2163 Ops/s $\textbf{\color{#35bf28}+17.99\%}$
test_redq_deprec_speed[True-backward] 10.2458ms 8.6061ms 116.1964 Ops/s 102.2744 Ops/s $\textbf{\color{#35bf28}+13.61\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.8573ms 3.9946ms 250.3358 Ops/s 208.5626 Ops/s $\textbf{\color{#35bf28}+20.03\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.9662ms 8.6467ms 115.6508 Ops/s 101.3840 Ops/s $\textbf{\color{#35bf28}+14.07\%}$
test_td3_speed[False-None] 10.9106ms 8.4139ms 118.8513 Ops/s 117.6712 Ops/s $\color{#35bf28}+1.00\%$
test_td3_speed[False-backward] 18.9943ms 11.1163ms 89.9581 Ops/s 86.7726 Ops/s $\color{#35bf28}+3.67\%$
test_td3_speed[True-None] 2.0392ms 1.8491ms 540.8018 Ops/s 515.9654 Ops/s $\color{#35bf28}+4.81\%$
test_td3_speed[True-backward] 3.6300ms 3.4845ms 286.9838 Ops/s 264.3587 Ops/s $\textbf{\color{#35bf28}+8.56\%}$
test_td3_speed[reduce-overhead-None] 2.1259ms 1.8306ms 546.2721 Ops/s 510.6741 Ops/s $\textbf{\color{#35bf28}+6.97\%}$
test_td3_speed[reduce-overhead-backward] 4.4320ms 3.5866ms 278.8184 Ops/s 266.5327 Ops/s $\color{#35bf28}+4.61\%$
test_cql_speed[False-None] 39.7132ms 37.1525ms 26.9161 Ops/s 26.3620 Ops/s $\color{#35bf28}+2.10\%$
test_cql_speed[False-backward] 56.8064ms 48.9659ms 20.4224 Ops/s 20.4917 Ops/s $\color{#d91a1a}-0.34\%$
test_cql_speed[True-None] 18.3010ms 16.7282ms 59.7791 Ops/s 58.0753 Ops/s $\color{#35bf28}+2.93\%$
test_cql_speed[True-backward] 24.9599ms 23.0939ms 43.3014 Ops/s 43.1455 Ops/s $\color{#35bf28}+0.36\%$
test_cql_speed[reduce-overhead-None] 18.0642ms 15.9983ms 62.5066 Ops/s 60.8068 Ops/s $\color{#35bf28}+2.80\%$
test_cql_speed[reduce-overhead-backward] 25.3101ms 23.8167ms 41.9874 Ops/s 42.4875 Ops/s $\color{#d91a1a}-1.18\%$
test_a2c_speed[False-None] 8.4089ms 7.3756ms 135.5822 Ops/s 136.8186 Ops/s $\color{#d91a1a}-0.90\%$
test_a2c_speed[False-backward] 15.6881ms 14.7842ms 67.6397 Ops/s 68.0490 Ops/s $\color{#d91a1a}-0.60\%$
test_a2c_speed[True-None] 5.6833ms 3.8050ms 262.8124 Ops/s 264.6230 Ops/s $\color{#d91a1a}-0.68\%$
test_a2c_speed[True-backward] 11.7870ms 11.1186ms 89.9393 Ops/s 98.2587 Ops/s $\textbf{\color{#d91a1a}-8.47\%}$
test_a2c_speed[reduce-overhead-None] 4.2509ms 3.8612ms 258.9853 Ops/s 267.0122 Ops/s $\color{#d91a1a}-3.01\%$
test_a2c_speed[reduce-overhead-backward] 11.3565ms 10.8539ms 92.1325 Ops/s 98.5511 Ops/s $\textbf{\color{#d91a1a}-6.51\%}$
test_ppo_speed[False-None] 8.7018ms 7.8246ms 127.8017 Ops/s 128.8546 Ops/s $\color{#d91a1a}-0.82\%$
test_ppo_speed[False-backward] 24.8902ms 16.0844ms 62.1719 Ops/s 66.3377 Ops/s $\textbf{\color{#d91a1a}-6.28\%}$
test_ppo_speed[True-None] 4.7395ms 4.2474ms 235.4370 Ops/s 239.0593 Ops/s $\color{#d91a1a}-1.52\%$
test_ppo_speed[True-backward] 11.4834ms 10.8608ms 92.0741 Ops/s 99.3473 Ops/s $\textbf{\color{#d91a1a}-7.32\%}$
test_ppo_speed[reduce-overhead-None] 4.8426ms 4.2504ms 235.2719 Ops/s 240.2346 Ops/s $\color{#d91a1a}-2.07\%$
test_ppo_speed[reduce-overhead-backward] 11.7380ms 10.4290ms 95.8861 Ops/s 99.5451 Ops/s $\color{#d91a1a}-3.68\%$
test_reinforce_speed[False-None] 8.0959ms 6.6308ms 150.8114 Ops/s 154.3375 Ops/s $\color{#d91a1a}-2.28\%$
test_reinforce_speed[False-backward] 10.0364ms 9.8498ms 101.5246 Ops/s 103.2380 Ops/s $\color{#d91a1a}-1.66\%$
test_reinforce_speed[True-None] 3.6509ms 3.1192ms 320.5948 Ops/s 322.2368 Ops/s $\color{#d91a1a}-0.51\%$
test_reinforce_speed[True-backward] 9.9854ms 9.1354ms 109.4645 Ops/s 110.3734 Ops/s $\color{#d91a1a}-0.82\%$
test_reinforce_speed[reduce-overhead-None] 7.3942ms 3.1835ms 314.1170 Ops/s 321.9758 Ops/s $\color{#d91a1a}-2.44\%$
test_reinforce_speed[reduce-overhead-backward] 10.1165ms 9.3032ms 107.4899 Ops/s 110.3404 Ops/s $\color{#d91a1a}-2.58\%$
test_iql_speed[False-None] 34.7075ms 33.0447ms 30.2620 Ops/s 30.3975 Ops/s $\color{#d91a1a}-0.45\%$
test_iql_speed[False-backward] 48.6235ms 46.5351ms 21.4892 Ops/s 22.0170 Ops/s $\color{#d91a1a}-2.40\%$
test_iql_speed[True-None] 12.9634ms 11.5546ms 86.5458 Ops/s 83.6757 Ops/s $\color{#35bf28}+3.43\%$
test_iql_speed[True-backward] 25.7788ms 23.9274ms 41.7931 Ops/s 42.7109 Ops/s $\color{#d91a1a}-2.15\%$
test_iql_speed[reduce-overhead-None] 12.2877ms 11.2944ms 88.5398 Ops/s 84.8204 Ops/s $\color{#35bf28}+4.39\%$
test_iql_speed[reduce-overhead-backward] 23.7110ms 22.8452ms 43.7728 Ops/s 44.3945 Ops/s $\color{#d91a1a}-1.40\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.6558ms 5.1677ms 193.5092 Ops/s 208.5591 Ops/s $\textbf{\color{#d91a1a}-7.22\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9488ms 0.5441ms 1.8378 KOps/s 1.8866 KOps/s $\color{#d91a1a}-2.59\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9077ms 0.5180ms 1.9305 KOps/s 1.9684 KOps/s $\color{#d91a1a}-1.92\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7948ms 4.8318ms 206.9637 Ops/s 220.9250 Ops/s $\textbf{\color{#d91a1a}-6.32\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.3441ms 0.5445ms 1.8366 KOps/s 1.9058 KOps/s $\color{#d91a1a}-3.63\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8047ms 0.5124ms 1.9516 KOps/s 1.8860 KOps/s $\color{#35bf28}+3.48\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.5906ms 1.7584ms 568.6875 Ops/s 590.7520 Ops/s $\color{#d91a1a}-3.73\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.2536ms 1.6719ms 598.1240 Ops/s 624.2422 Ops/s $\color{#d91a1a}-4.18\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.5272ms 4.8653ms 205.5391 Ops/s 204.9534 Ops/s $\color{#35bf28}+0.29\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2242ms 0.6799ms 1.4708 KOps/s 1.4999 KOps/s $\color{#d91a1a}-1.94\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0547ms 0.6565ms 1.5232 KOps/s 1.5592 KOps/s $\color{#d91a1a}-2.31\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.1879ms 4.8364ms 206.7647 Ops/s 207.9207 Ops/s $\color{#d91a1a}-0.56\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2379ms 0.5447ms 1.8360 KOps/s 1.8568 KOps/s $\color{#d91a1a}-1.12\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7504ms 0.5125ms 1.9511 KOps/s 1.9013 KOps/s $\color{#35bf28}+2.62\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.4211ms 4.7155ms 212.0671 Ops/s 205.2913 Ops/s $\color{#35bf28}+3.30\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.3220ms 0.5361ms 1.8653 KOps/s 1.8813 KOps/s $\color{#d91a1a}-0.85\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8871ms 0.5243ms 1.9073 KOps/s 1.9647 KOps/s $\color{#d91a1a}-2.92\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.7626ms 5.0388ms 198.4592 Ops/s 197.5221 Ops/s $\color{#35bf28}+0.47\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1499ms 0.6886ms 1.4523 KOps/s 1.4653 KOps/s $\color{#d91a1a}-0.89\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8960ms 0.6602ms 1.5148 KOps/s 1.5442 KOps/s $\color{#d91a1a}-1.91\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.7652ms 4.3165ms 231.6695 Ops/s 232.4592 Ops/s $\color{#d91a1a}-0.34\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.3905ms 2.5024ms 399.6109 Ops/s 387.4438 Ops/s $\color{#35bf28}+3.14\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 3.9663ms 1.3520ms 739.6406 Ops/s 694.4688 Ops/s $\textbf{\color{#35bf28}+6.50\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 5.6399ms 4.2777ms 233.7722 Ops/s 236.2235 Ops/s $\color{#d91a1a}-1.04\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.4675s 11.8101ms 84.6735 Ops/s 419.8881 Ops/s $\textbf{\color{#d91a1a}-79.83\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.4440ms 1.4089ms 709.7882 Ops/s 699.1664 Ops/s $\color{#35bf28}+1.52\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 5.8110ms 4.4928ms 222.5802 Ops/s 237.4177 Ops/s $\textbf{\color{#d91a1a}-6.25\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.4183ms 2.4934ms 401.0558 Ops/s 394.1345 Ops/s $\color{#35bf28}+1.76\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.1807ms 1.5578ms 641.9335 Ops/s 606.5720 Ops/s $\textbf{\color{#35bf28}+5.83\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 12.4915ms 12.0715ms 82.8397 Ops/s 75.4958 Ops/s $\textbf{\color{#35bf28}+9.73\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.0306ms 14.6621ms 68.2031 Ops/s 70.3924 Ops/s $\color{#d91a1a}-3.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 23.5881ms 20.9025ms 47.8412 Ops/s 45.9849 Ops/s $\color{#35bf28}+4.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.2231ms 14.8106ms 67.5192 Ops/s 67.3249 Ops/s $\color{#35bf28}+0.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 22.1332ms 20.8051ms 48.0652 Ops/s 46.2444 Ops/s $\color{#35bf28}+3.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 18.9656ms 16.2355ms 61.5935 Ops/s 60.8829 Ops/s $\color{#35bf28}+1.17\%$

Copy link

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 149. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.9187s 0.8275s 1.2085 Ops/s 1.2303 Ops/s $\color{#d91a1a}-1.78\%$
test_transformed 1.5502s 1.4616s 0.6842 Ops/s 0.6739 Ops/s $\color{#35bf28}+1.52\%$
test_serial 2.4324s 2.3405s 0.4273 Ops/s 0.4217 Ops/s $\color{#35bf28}+1.31\%$
test_parallel 1.9278s 1.8389s 0.5438 Ops/s 0.5345 Ops/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[True-True-True-True-True] 0.2134ms 40.6928μs 24.5743 KOps/s 24.8977 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[True-True-True-True-False] 0.1284ms 23.2661μs 42.9811 KOps/s 42.5853 KOps/s $\color{#35bf28}+0.93\%$
test_step_mdp_speed[True-True-True-False-True] 58.5210μs 22.4378μs 44.5676 KOps/s 44.8808 KOps/s $\color{#d91a1a}-0.70\%$
test_step_mdp_speed[True-True-True-False-False] 47.7810μs 13.0587μs 76.5774 KOps/s 76.6730 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-True-False-True-True] 80.8820μs 42.7233μs 23.4064 KOps/s 23.5324 KOps/s $\color{#d91a1a}-0.54\%$
test_step_mdp_speed[True-True-False-True-False] 59.2110μs 25.4312μs 39.3217 KOps/s 38.8524 KOps/s $\color{#35bf28}+1.21\%$
test_step_mdp_speed[True-True-False-False-True] 52.0210μs 24.5848μs 40.6755 KOps/s 42.3028 KOps/s $\color{#d91a1a}-3.85\%$
test_step_mdp_speed[True-True-False-False-False] 42.3200μs 15.3034μs 65.3448 KOps/s 65.5133 KOps/s $\color{#d91a1a}-0.26\%$
test_step_mdp_speed[True-False-True-True-True] 78.2810μs 44.6401μs 22.4014 KOps/s 22.4072 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[True-False-True-True-False] 93.7010μs 27.9243μs 35.8111 KOps/s 35.8093 KOps/s $+0.01\%$
test_step_mdp_speed[True-False-True-False-True] 52.7910μs 24.4635μs 40.8772 KOps/s 41.0596 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[True-False-True-False-False] 47.8610μs 14.9977μs 66.6769 KOps/s 65.8562 KOps/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-False-False-True-True] 80.7110μs 46.0731μs 21.7046 KOps/s 21.3308 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[True-False-False-True-False] 59.2810μs 29.9332μs 33.4077 KOps/s 33.1394 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[True-False-False-False-True] 0.1013ms 26.5028μs 37.7319 KOps/s 37.4815 KOps/s $\color{#35bf28}+0.67\%$
test_step_mdp_speed[True-False-False-False-False] 0.1304ms 17.7570μs 56.3159 KOps/s 56.8549 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[False-True-True-True-True] 77.4110μs 44.7985μs 22.3222 KOps/s 22.2521 KOps/s $\color{#35bf28}+0.31\%$
test_step_mdp_speed[False-True-True-True-False] 61.0810μs 28.0734μs 35.6209 KOps/s 35.6161 KOps/s $\color{#35bf28}+0.01\%$
test_step_mdp_speed[False-True-True-False-True] 88.9310μs 28.3547μs 35.2675 KOps/s 35.2300 KOps/s $\color{#35bf28}+0.11\%$
test_step_mdp_speed[False-True-True-False-False] 42.1710μs 16.9362μs 59.0452 KOps/s 58.6825 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[False-True-False-True-True] 72.1910μs 46.6852μs 21.4201 KOps/s 21.3878 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[False-True-False-True-False] 57.9010μs 30.2290μs 33.0809 KOps/s 32.9260 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[False-True-False-False-True] 3.6323ms 30.6747μs 32.6001 KOps/s 33.2121 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[False-True-False-False-False] 0.1563ms 19.2406μs 51.9734 KOps/s 53.2165 KOps/s $\color{#d91a1a}-2.34\%$
test_step_mdp_speed[False-False-True-True-True] 94.5010μs 48.8755μs 20.4602 KOps/s 20.3192 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[False-False-True-True-False] 55.9810μs 32.6575μs 30.6208 KOps/s 30.5652 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-False-True-False-True] 58.4810μs 30.5802μs 32.7009 KOps/s 32.3532 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-False-True-False-False] 45.7210μs 19.2591μs 51.9235 KOps/s 51.6015 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[False-False-False-True-True] 87.5910μs 51.3469μs 19.4754 KOps/s 19.3982 KOps/s $\color{#35bf28}+0.40\%$
test_step_mdp_speed[False-False-False-True-False] 61.4100μs 34.9639μs 28.6010 KOps/s 28.3877 KOps/s $\color{#35bf28}+0.75\%$
test_step_mdp_speed[False-False-False-False-True] 65.6610μs 32.1716μs 31.0833 KOps/s 31.2632 KOps/s $\color{#d91a1a}-0.58\%$
test_step_mdp_speed[False-False-False-False-False] 58.9810μs 21.4009μs 46.7271 KOps/s 46.9406 KOps/s $\color{#d91a1a}-0.45\%$
test_values[generalized_advantage_estimate-True-True] 26.1084ms 25.4340ms 39.3175 Ops/s 39.8906 Ops/s $\color{#d91a1a}-1.44\%$
test_values[vec_generalized_advantage_estimate-True-True] 99.8559ms 2.9002ms 344.8096 Ops/s 325.7819 Ops/s $\textbf{\color{#35bf28}+5.84\%}$
test_values[td0_return_estimate-False-False] 0.1073ms 80.5464μs 12.4152 KOps/s 12.5622 KOps/s $\color{#d91a1a}-1.17\%$
test_values[td1_return_estimate-False-False] 56.5471ms 55.7389ms 17.9408 Ops/s 18.1702 Ops/s $\color{#d91a1a}-1.26\%$
test_values[vec_td1_return_estimate-False-False] 1.2692ms 1.0943ms 913.8554 Ops/s 922.7675 Ops/s $\color{#d91a1a}-0.97\%$
test_values[td_lambda_return_estimate-True-False] 89.2893ms 88.4436ms 11.3066 Ops/s 11.4482 Ops/s $\color{#d91a1a}-1.24\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2526ms 1.0843ms 922.2368 Ops/s 932.8148 Ops/s $\color{#d91a1a}-1.13\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.7128ms 25.3797ms 39.4015 Ops/s 40.3616 Ops/s $\color{#d91a1a}-2.38\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0597ms 0.7596ms 1.3165 KOps/s 1.3217 KOps/s $\color{#d91a1a}-0.39\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8235ms 0.6737ms 1.4843 KOps/s 1.4982 KOps/s $\color{#d91a1a}-0.93\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6338ms 1.4892ms 671.5155 Ops/s 675.5192 Ops/s $\color{#d91a1a}-0.59\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8496ms 0.6890ms 1.4514 KOps/s 1.4665 KOps/s $\color{#d91a1a}-1.03\%$
test_dqn_speed[False-None] 7.3419ms 1.5440ms 647.6689 Ops/s 643.6434 Ops/s $\color{#35bf28}+0.63\%$
test_dqn_speed[False-backward] 2.2906ms 2.1474ms 465.6692 Ops/s 457.0920 Ops/s $\color{#35bf28}+1.88\%$
test_dqn_speed[True-None] 0.9939ms 0.5702ms 1.7537 KOps/s 1.7300 KOps/s $\color{#35bf28}+1.37\%$
test_dqn_speed[True-backward] 1.3828ms 1.2135ms 824.0580 Ops/s 790.2797 Ops/s $\color{#35bf28}+4.27\%$
test_dqn_speed[reduce-overhead-None] 1.0276ms 0.5988ms 1.6699 KOps/s 1.6949 KOps/s $\color{#d91a1a}-1.48\%$
test_dqn_speed[reduce-overhead-backward] 1.2442ms 1.0934ms 914.5372 Ops/s 960.9087 Ops/s $\color{#d91a1a}-4.83\%$
test_ddpg_speed[False-None] 3.3536ms 2.9185ms 342.6430 Ops/s 338.3218 Ops/s $\color{#35bf28}+1.28\%$
test_ddpg_speed[False-backward] 4.3974ms 4.1684ms 239.9022 Ops/s 233.6373 Ops/s $\color{#35bf28}+2.68\%$
test_ddpg_speed[True-None] 1.7967ms 1.3736ms 728.0291 Ops/s 727.6651 Ops/s $\color{#35bf28}+0.05\%$
test_ddpg_speed[True-backward] 2.9130ms 2.5323ms 394.8933 Ops/s 379.2659 Ops/s $\color{#35bf28}+4.12\%$
test_ddpg_speed[reduce-overhead-None] 1.8704ms 1.3948ms 716.9648 Ops/s 719.8269 Ops/s $\color{#d91a1a}-0.40\%$
test_ddpg_speed[reduce-overhead-backward] 2.0685ms 1.9282ms 518.6177 Ops/s 477.3990 Ops/s $\textbf{\color{#35bf28}+8.63\%}$
test_sac_speed[False-None] 8.5651ms 8.1468ms 122.7478 Ops/s 121.4016 Ops/s $\color{#35bf28}+1.11\%$
test_sac_speed[False-backward] 11.7249ms 11.1685ms 89.5378 Ops/s 87.3024 Ops/s $\color{#35bf28}+2.56\%$
test_sac_speed[True-None] 2.0744ms 1.8929ms 528.2854 Ops/s 519.1672 Ops/s $\color{#35bf28}+1.76\%$
test_sac_speed[True-backward] 4.0316ms 3.8323ms 260.9377 Ops/s 270.3593 Ops/s $\color{#d91a1a}-3.48\%$
test_sac_speed[reduce-overhead-None] 20.6548ms 11.8331ms 84.5085 Ops/s 82.0988 Ops/s $\color{#35bf28}+2.94\%$
test_sac_speed[reduce-overhead-backward] 1.9989ms 1.8500ms 540.5471 Ops/s 576.4260 Ops/s $\textbf{\color{#d91a1a}-6.22\%}$
test_redq_speed[False-None] 8.7835ms 7.5483ms 132.4797 Ops/s 126.2277 Ops/s $\color{#35bf28}+4.95\%$
test_redq_speed[False-backward] 11.9167ms 11.5252ms 86.7663 Ops/s 85.2957 Ops/s $\color{#35bf28}+1.72\%$
test_redq_speed[True-None] 2.5819ms 2.3780ms 420.5142 Ops/s 415.9272 Ops/s $\color{#35bf28}+1.10\%$
test_redq_speed[True-backward] 4.3782ms 4.2443ms 235.6093 Ops/s 238.8887 Ops/s $\color{#d91a1a}-1.37\%$
test_redq_speed[reduce-overhead-None] 2.5764ms 2.4008ms 416.5318 Ops/s 408.7448 Ops/s $\color{#35bf28}+1.91\%$
test_redq_speed[reduce-overhead-backward] 4.4158ms 4.2600ms 234.7432 Ops/s 230.9919 Ops/s $\color{#35bf28}+1.62\%$
test_redq_deprec_speed[False-None] 9.5015ms 9.1326ms 109.4975 Ops/s 109.1332 Ops/s $\color{#35bf28}+0.33\%$
test_redq_deprec_speed[False-backward] 12.9195ms 12.4444ms 80.3573 Ops/s 79.9774 Ops/s $\color{#35bf28}+0.48\%$
test_redq_deprec_speed[True-None] 2.8985ms 2.7070ms 369.4135 Ops/s 365.4184 Ops/s $\color{#35bf28}+1.09\%$
test_redq_deprec_speed[True-backward] 4.8061ms 4.5798ms 218.3512 Ops/s 216.6499 Ops/s $\color{#35bf28}+0.79\%$
test_redq_deprec_speed[reduce-overhead-None] 2.8445ms 2.6768ms 373.5845 Ops/s 365.5382 Ops/s $\color{#35bf28}+2.20\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.9484ms 4.5458ms 219.9843 Ops/s 215.4557 Ops/s $\color{#35bf28}+2.10\%$
test_td3_speed[False-None] 8.2238ms 8.0318ms 124.5054 Ops/s 123.6271 Ops/s $\color{#35bf28}+0.71\%$
test_td3_speed[False-backward] 11.2479ms 10.6145ms 94.2110 Ops/s 94.1418 Ops/s $\color{#35bf28}+0.07\%$
test_td3_speed[True-None] 1.7992ms 1.7188ms 581.7931 Ops/s 547.6626 Ops/s $\textbf{\color{#35bf28}+6.23\%}$
test_td3_speed[True-backward] 3.5753ms 3.4219ms 292.2391 Ops/s 288.3517 Ops/s $\color{#35bf28}+1.35\%$
test_td3_speed[reduce-overhead-None] 51.9190ms 26.3780ms 37.9104 Ops/s 36.6476 Ops/s $\color{#35bf28}+3.45\%$
test_td3_speed[reduce-overhead-backward] 1.7022ms 1.5574ms 642.0838 Ops/s 627.2073 Ops/s $\color{#35bf28}+2.37\%$
test_cql_speed[False-None] 17.6137ms 16.9857ms 58.8731 Ops/s 57.4959 Ops/s $\color{#35bf28}+2.40\%$
test_cql_speed[False-backward] 23.7239ms 22.5969ms 44.2539 Ops/s 42.9867 Ops/s $\color{#35bf28}+2.95\%$
test_cql_speed[True-None] 3.7864ms 3.5748ms 279.7364 Ops/s 283.8043 Ops/s $\color{#d91a1a}-1.43\%$
test_cql_speed[True-backward] 6.0873ms 5.6181ms 177.9948 Ops/s 177.2427 Ops/s $\color{#35bf28}+0.42\%$
test_cql_speed[reduce-overhead-None] 21.0240ms 13.2042ms 75.7337 Ops/s 74.3951 Ops/s $\color{#35bf28}+1.80\%$
test_cql_speed[reduce-overhead-backward] 2.2602ms 2.0547ms 486.6782 Ops/s 519.6255 Ops/s $\textbf{\color{#d91a1a}-6.34\%}$
test_a2c_speed[False-None] 3.4726ms 3.2048ms 312.0301 Ops/s 304.4470 Ops/s $\color{#35bf28}+2.49\%$
test_a2c_speed[False-backward] 6.9792ms 6.3710ms 156.9621 Ops/s 159.0334 Ops/s $\color{#d91a1a}-1.30\%$
test_a2c_speed[True-None] 1.5433ms 1.3770ms 726.2169 Ops/s 718.2849 Ops/s $\color{#35bf28}+1.10\%$
test_a2c_speed[True-backward] 3.2891ms 3.0947ms 323.1327 Ops/s 319.3081 Ops/s $\color{#35bf28}+1.20\%$
test_a2c_speed[reduce-overhead-None] 15.9006ms 9.0238ms 110.8177 Ops/s 112.1238 Ops/s $\color{#d91a1a}-1.16\%$
test_a2c_speed[reduce-overhead-backward] 1.7770ms 1.6434ms 608.5079 Ops/s 602.2013 Ops/s $\color{#35bf28}+1.05\%$
test_ppo_speed[False-None] 3.8974ms 3.7127ms 269.3451 Ops/s 262.9071 Ops/s $\color{#35bf28}+2.45\%$
test_ppo_speed[False-backward] 7.5545ms 7.0877ms 141.0891 Ops/s 139.2097 Ops/s $\color{#35bf28}+1.35\%$
test_ppo_speed[True-None] 1.6123ms 1.4437ms 692.6614 Ops/s 684.3787 Ops/s $\color{#35bf28}+1.21\%$
test_ppo_speed[True-backward] 3.3954ms 3.2407ms 308.5756 Ops/s 303.4896 Ops/s $\color{#35bf28}+1.68\%$
test_ppo_speed[reduce-overhead-None] 1.1669ms 0.9938ms 1.0062 KOps/s 1.0087 KOps/s $\color{#d91a1a}-0.25\%$
test_ppo_speed[reduce-overhead-backward] 1.7676ms 1.6029ms 623.8754 Ops/s 602.1600 Ops/s $\color{#35bf28}+3.61\%$
test_reinforce_speed[False-None] 2.4854ms 2.2961ms 435.5274 Ops/s 426.7869 Ops/s $\color{#35bf28}+2.05\%$
test_reinforce_speed[False-backward] 3.8718ms 3.4589ms 289.1095 Ops/s 285.4657 Ops/s $\color{#35bf28}+1.28\%$
test_reinforce_speed[True-None] 1.5034ms 1.3260ms 754.1670 Ops/s 748.6410 Ops/s $\color{#35bf28}+0.74\%$
test_reinforce_speed[True-backward] 3.2265ms 3.1019ms 322.3815 Ops/s 323.0176 Ops/s $\color{#d91a1a}-0.20\%$
test_reinforce_speed[reduce-overhead-None] 17.6540ms 9.9208ms 100.7979 Ops/s 99.8801 Ops/s $\color{#35bf28}+0.92\%$
test_reinforce_speed[reduce-overhead-backward] 1.7724ms 1.6805ms 595.0627 Ops/s 580.9303 Ops/s $\color{#35bf28}+2.43\%$
test_iql_speed[False-None] 9.7625ms 9.2847ms 107.7037 Ops/s 105.5405 Ops/s $\color{#35bf28}+2.05\%$
test_iql_speed[False-backward] 13.9345ms 13.2894ms 75.2482 Ops/s 74.1413 Ops/s $\color{#35bf28}+1.49\%$
test_iql_speed[True-None] 2.5055ms 2.2960ms 435.5311 Ops/s 404.2819 Ops/s $\textbf{\color{#35bf28}+7.73\%}$
test_iql_speed[True-backward] 5.0903ms 4.9528ms 201.9063 Ops/s 196.9541 Ops/s $\color{#35bf28}+2.51\%$
test_iql_speed[reduce-overhead-None] 0.5004s 12.9585ms 77.1694 Ops/s 88.3565 Ops/s $\textbf{\color{#d91a1a}-12.66\%}$
test_iql_speed[reduce-overhead-backward] 2.2902ms 2.1258ms 470.4107 Ops/s 481.0576 Ops/s $\color{#d91a1a}-2.21\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7960ms 6.3228ms 158.1584 Ops/s 155.4400 Ops/s $\color{#35bf28}+1.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6382ms 0.3533ms 2.8303 KOps/s 3.4854 KOps/s $\textbf{\color{#d91a1a}-18.80\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6177ms 0.3328ms 3.0047 KOps/s 3.8334 KOps/s $\textbf{\color{#d91a1a}-21.62\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3827ms 6.0633ms 164.9274 Ops/s 162.2996 Ops/s $\color{#35bf28}+1.62\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1077ms 0.3184ms 3.1411 KOps/s 2.7972 KOps/s $\textbf{\color{#35bf28}+12.29\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5797ms 0.2970ms 3.3670 KOps/s 3.9973 KOps/s $\textbf{\color{#d91a1a}-15.77\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.8951ms 1.4526ms 688.4316 Ops/s 747.7897 Ops/s $\textbf{\color{#d91a1a}-7.94\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5745ms 1.3405ms 745.9691 Ops/s 768.0044 Ops/s $\color{#d91a1a}-2.87\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5152ms 6.2374ms 160.3227 Ops/s 154.9039 Ops/s $\color{#35bf28}+3.50\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1529ms 0.4344ms 2.3022 KOps/s 1.9498 KOps/s $\textbf{\color{#35bf28}+18.08\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7631ms 0.4682ms 2.1360 KOps/s 2.0775 KOps/s $\color{#35bf28}+2.82\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2544ms 6.0602ms 165.0110 Ops/s 161.7060 Ops/s $\color{#35bf28}+2.04\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8465ms 0.3573ms 2.7986 KOps/s 2.8736 KOps/s $\color{#d91a1a}-2.61\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5935ms 0.3420ms 2.9239 KOps/s 3.0302 KOps/s $\color{#d91a1a}-3.51\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.5188ms 6.0427ms 165.4876 Ops/s 162.5352 Ops/s $\color{#35bf28}+1.82\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6185ms 0.3471ms 2.8810 KOps/s 2.8481 KOps/s $\color{#35bf28}+1.16\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5794ms 0.3086ms 3.2409 KOps/s 3.3892 KOps/s $\color{#d91a1a}-4.38\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4495ms 6.2679ms 159.5439 Ops/s 155.6597 Ops/s $\color{#35bf28}+2.50\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2353ms 0.5013ms 1.9947 KOps/s 2.0620 KOps/s $\color{#d91a1a}-3.26\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6937ms 0.4294ms 2.3289 KOps/s 2.3828 KOps/s $\color{#d91a1a}-2.26\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.3063ms 5.5010ms 181.7845 Ops/s 175.3142 Ops/s $\color{#35bf28}+3.69\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 5.6605ms 2.0259ms 493.6150 Ops/s 425.3254 Ops/s $\textbf{\color{#35bf28}+16.06\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 10.0154ms 1.3023ms 767.8670 Ops/s 812.2866 Ops/s $\textbf{\color{#d91a1a}-5.47\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4705s 14.9067ms 67.0841 Ops/s 177.2772 Ops/s $\textbf{\color{#d91a1a}-62.16\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.8579ms 1.8095ms 552.6479 Ops/s 427.8317 Ops/s $\textbf{\color{#35bf28}+29.17\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.6615ms 1.2672ms 789.1211 Ops/s 840.4909 Ops/s $\textbf{\color{#d91a1a}-6.11\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.6716ms 5.7606ms 173.5924 Ops/s 30.0614 Ops/s $\textbf{\color{#35bf28}+477.46\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.8814ms 2.2280ms 448.8369 Ops/s 526.9355 Ops/s $\textbf{\color{#d91a1a}-14.82\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.3071ms 1.4445ms 692.2647 Ops/s 800.9168 Ops/s $\textbf{\color{#d91a1a}-13.57\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.5522ms 13.2212ms 75.6360 Ops/s 71.5937 Ops/s $\textbf{\color{#35bf28}+5.65\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.7157ms 16.9414ms 59.0271 Ops/s 58.5407 Ops/s $\color{#35bf28}+0.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.2895ms 17.9982ms 55.5611 Ops/s 53.8314 Ops/s $\color{#35bf28}+3.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7675ms 17.1366ms 58.3547 Ops/s 57.9520 Ops/s $\color{#35bf28}+0.69\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 0.4070s 25.9234ms 38.5751 Ops/s 54.4080 Ops/s $\textbf{\color{#d91a1a}-29.10\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.2012ms 18.6363ms 53.6586 Ops/s 54.1491 Ops/s $\color{#d91a1a}-0.91\%$

@vmoens vmoens merged commit 90c51f7 into gh/vmoens/92/base Feb 20, 2025
62 of 74 checks passed
vmoens added a commit that referenced this pull request Feb 20, 2025
ghstack-source-id: 59925635a87b196a5bcb0fb251afe4cc7b8b103e
Pull Request resolved: #2798
@vmoens vmoens deleted the gh/vmoens/92/head branch February 20, 2025 21:24
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants