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] Ensure that Composite.set returns self as TensorDict does #2784

Open
wants to merge 2 commits into
base: gh/vmoens/89/base
Choose a base branch
from

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 12, 2025

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 12, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2784

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

❌ 11 New Failures

As of commit 21d5954 with merge base b27ee6d (image):

NEW FAILURES - The following jobs have failed:

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

@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 12, 2025
vmoens added a commit that referenced this pull request Feb 12, 2025
ghstack-source-id: ee47d30c335a95b5a100ba4a32f10c578cacdbdc
Pull Request resolved: #2784
Copy link

github-actions bot commented Feb 12, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5894s 0.4996s 2.0017 Ops/s 1.9549 Ops/s $\color{#35bf28}+2.39\%$
test_transformed 1.0385s 0.9528s 1.0495 Ops/s 0.9986 Ops/s $\textbf{\color{#35bf28}+5.09\%}$
test_serial 1.5764s 1.4852s 0.6733 Ops/s 0.6603 Ops/s $\color{#35bf28}+1.98\%$
test_parallel 1.3769s 1.3004s 0.7690 Ops/s 0.7625 Ops/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[True-True-True-True-True] 0.3334ms 29.9582μs 33.3798 KOps/s 32.9681 KOps/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-True-True-True-False] 54.2010μs 17.9794μs 55.6192 KOps/s 56.1762 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[True-True-True-False-True] 48.3400μs 17.1550μs 58.2920 KOps/s 58.2440 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[True-True-True-False-False] 42.5590μs 10.1803μs 98.2288 KOps/s 99.1904 KOps/s $\color{#d91a1a}-0.97\%$
test_step_mdp_speed[True-True-False-True-True] 82.9640μs 32.3554μs 30.9067 KOps/s 30.9367 KOps/s $\color{#d91a1a}-0.10\%$
test_step_mdp_speed[True-True-False-True-False] 51.4450μs 19.7426μs 50.6519 KOps/s 50.8724 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[True-True-False-False-True] 47.9790μs 18.9661μs 52.7255 KOps/s 52.7049 KOps/s $\color{#35bf28}+0.04\%$
test_step_mdp_speed[True-True-False-False-False] 58.9390μs 12.0793μs 82.7860 KOps/s 83.7229 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[True-False-True-True-True] 74.4880μs 34.2896μs 29.1634 KOps/s 29.2379 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[True-False-True-True-False] 71.8130μs 21.6548μs 46.1791 KOps/s 46.0042 KOps/s $\color{#35bf28}+0.38\%$
test_step_mdp_speed[True-False-True-False-True] 66.1130μs 18.9655μs 52.7273 KOps/s 52.9463 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[True-False-True-False-False] 40.4650μs 12.0758μs 82.8104 KOps/s 84.3853 KOps/s $\color{#d91a1a}-1.87\%$
test_step_mdp_speed[True-False-False-True-True] 88.4540μs 35.5803μs 28.1054 KOps/s 27.9642 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-False-False-True-False] 84.2550μs 23.0822μs 43.3234 KOps/s 42.8359 KOps/s $\color{#35bf28}+1.14\%$
test_step_mdp_speed[True-False-False-False-True] 0.5857ms 20.5386μs 48.6888 KOps/s 48.7442 KOps/s $\color{#d91a1a}-0.11\%$
test_step_mdp_speed[True-False-False-False-False] 62.1350μs 13.6182μs 73.4313 KOps/s 73.2133 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[False-True-True-True-True] 93.4120μs 33.7498μs 29.6298 KOps/s 29.1941 KOps/s $\color{#35bf28}+1.49\%$
test_step_mdp_speed[False-True-True-True-False] 68.1290μs 21.5329μs 46.4407 KOps/s 45.7423 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-True-True-False-True] 71.9740μs 21.8239μs 45.8214 KOps/s 46.9493 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[False-True-True-False-False] 43.5100μs 13.3290μs 75.0242 KOps/s 75.5887 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-True-False-True-True] 89.4650μs 35.5383μs 28.1387 KOps/s 28.3466 KOps/s $\color{#d91a1a}-0.73\%$
test_step_mdp_speed[False-True-False-True-False] 56.4140μs 23.1501μs 43.1963 KOps/s 42.8896 KOps/s $\color{#35bf28}+0.72\%$
test_step_mdp_speed[False-True-False-False-True] 2.6007ms 23.5389μs 42.4830 KOps/s 42.0939 KOps/s $\color{#35bf28}+0.92\%$
test_step_mdp_speed[False-True-False-False-False] 56.1440μs 15.1030μs 66.2121 KOps/s 66.8518 KOps/s $\color{#d91a1a}-0.96\%$
test_step_mdp_speed[False-False-True-True-True] 81.3810μs 37.6478μs 26.5620 KOps/s 26.4904 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[False-False-True-True-False] 76.4220μs 25.1704μs 39.7291 KOps/s 39.5417 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[False-False-True-False-True] 64.6500μs 23.4686μs 42.6102 KOps/s 43.3259 KOps/s $\color{#d91a1a}-1.65\%$
test_step_mdp_speed[False-False-True-False-False] 39.8240μs 15.1909μs 65.8291 KOps/s 66.8748 KOps/s $\color{#d91a1a}-1.56\%$
test_step_mdp_speed[False-False-False-True-True] 89.0060μs 39.5287μs 25.2981 KOps/s 25.3817 KOps/s $\color{#d91a1a}-0.33\%$
test_step_mdp_speed[False-False-False-True-False] 78.6360μs 26.9143μs 37.1549 KOps/s 37.0507 KOps/s $\color{#35bf28}+0.28\%$
test_step_mdp_speed[False-False-False-False-True] 0.5901ms 25.0624μs 39.9004 KOps/s 40.0354 KOps/s $\color{#d91a1a}-0.34\%$
test_step_mdp_speed[False-False-False-False-False] 50.4830μs 16.8546μs 59.3308 KOps/s 60.0747 KOps/s $\color{#d91a1a}-1.24\%$
test_values[generalized_advantage_estimate-True-True] 12.8059ms 9.9462ms 100.5410 Ops/s 101.2426 Ops/s $\color{#d91a1a}-0.69\%$
test_values[vec_generalized_advantage_estimate-True-True] 28.1071ms 25.9936ms 38.4711 Ops/s 38.6863 Ops/s $\color{#d91a1a}-0.56\%$
test_values[td0_return_estimate-False-False] 0.2467ms 0.1769ms 5.6530 KOps/s 5.5479 KOps/s $\color{#35bf28}+1.89\%$
test_values[td1_return_estimate-False-False] 27.6187ms 24.0495ms 41.5809 Ops/s 41.8268 Ops/s $\color{#d91a1a}-0.59\%$
test_values[vec_td1_return_estimate-False-False] 27.9954ms 26.0107ms 38.4456 Ops/s 38.2562 Ops/s $\color{#35bf28}+0.50\%$
test_values[td_lambda_return_estimate-True-False] 35.5862ms 34.7682ms 28.7619 Ops/s 29.1592 Ops/s $\color{#d91a1a}-1.36\%$
test_values[vec_td_lambda_return_estimate-True-False] 29.6498ms 26.1692ms 38.2128 Ops/s 38.0843 Ops/s $\color{#35bf28}+0.34\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.2218ms 8.6102ms 116.1407 Ops/s 116.5234 Ops/s $\color{#d91a1a}-0.33\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.2212ms 1.9889ms 502.8001 Ops/s 542.6707 Ops/s $\textbf{\color{#d91a1a}-7.35\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5135ms 0.3659ms 2.7331 KOps/s 2.7170 KOps/s $\color{#35bf28}+0.59\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 50.3417ms 45.6112ms 21.9245 Ops/s 21.6829 Ops/s $\color{#35bf28}+1.11\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.2202ms 3.4290ms 291.6265 Ops/s 289.8723 Ops/s $\color{#35bf28}+0.61\%$
test_dqn_speed[False-None] 6.1742ms 1.4114ms 708.5371 Ops/s 674.5363 Ops/s $\textbf{\color{#35bf28}+5.04\%}$
test_dqn_speed[False-backward] 2.0145ms 1.8915ms 528.6820 Ops/s 521.0368 Ops/s $\color{#35bf28}+1.47\%$
test_dqn_speed[True-None] 0.7582ms 0.4891ms 2.0447 KOps/s 2.0268 KOps/s $\color{#35bf28}+0.88\%$
test_dqn_speed[True-backward] 0.9854ms 0.9220ms 1.0846 KOps/s 789.5745 Ops/s $\textbf{\color{#35bf28}+37.36\%}$
test_dqn_speed[reduce-overhead-None] 0.6336ms 0.4891ms 2.0444 KOps/s 2.0441 KOps/s $\color{#35bf28}+0.01\%$
test_dqn_speed[reduce-overhead-backward] 0.9912ms 0.9101ms 1.0988 KOps/s 1.0934 KOps/s $\color{#35bf28}+0.50\%$
test_ddpg_speed[False-None] 3.2445ms 2.9119ms 343.4236 Ops/s 337.4735 Ops/s $\color{#35bf28}+1.76\%$
test_ddpg_speed[False-backward] 4.1702ms 4.0483ms 247.0186 Ops/s 245.9642 Ops/s $\color{#35bf28}+0.43\%$
test_ddpg_speed[True-None] 1.4779ms 1.2406ms 806.0715 Ops/s 805.0066 Ops/s $\color{#35bf28}+0.13\%$
test_ddpg_speed[True-backward] 2.2466ms 2.1582ms 463.3400 Ops/s 467.6120 Ops/s $\color{#d91a1a}-0.91\%$
test_ddpg_speed[reduce-overhead-None] 1.4113ms 1.2440ms 803.8800 Ops/s 800.8424 Ops/s $\color{#35bf28}+0.38\%$
test_ddpg_speed[reduce-overhead-backward] 2.2054ms 2.1350ms 468.3771 Ops/s 468.8111 Ops/s $\color{#d91a1a}-0.09\%$
test_sac_speed[False-None] 9.3894ms 8.1542ms 122.6358 Ops/s 121.6152 Ops/s $\color{#35bf28}+0.84\%$
test_sac_speed[False-backward] 13.1666ms 11.0974ms 90.1114 Ops/s 91.5000 Ops/s $\color{#d91a1a}-1.52\%$
test_sac_speed[True-None] 2.7640ms 2.1088ms 474.2104 Ops/s 473.1877 Ops/s $\color{#35bf28}+0.22\%$
test_sac_speed[True-backward] 5.0605ms 3.8396ms 260.4438 Ops/s 261.2204 Ops/s $\color{#d91a1a}-0.30\%$
test_sac_speed[reduce-overhead-None] 2.8078ms 2.1492ms 465.2854 Ops/s 467.2835 Ops/s $\color{#d91a1a}-0.43\%$
test_sac_speed[reduce-overhead-backward] 3.9822ms 3.8625ms 258.8987 Ops/s 258.6146 Ops/s $\color{#35bf28}+0.11\%$
test_redq_speed[False-None] 13.9972ms 13.3568ms 74.8682 Ops/s 70.3572 Ops/s $\textbf{\color{#35bf28}+6.41\%}$
test_redq_speed[False-backward] 23.7964ms 22.5241ms 44.3969 Ops/s 41.4436 Ops/s $\textbf{\color{#35bf28}+7.13\%}$
test_redq_speed[True-None] 7.0099ms 5.8230ms 171.7328 Ops/s 190.9678 Ops/s $\textbf{\color{#d91a1a}-10.07\%}$
test_redq_speed[True-backward] 14.1011ms 13.5416ms 73.8463 Ops/s 77.5216 Ops/s $\color{#d91a1a}-4.74\%$
test_redq_speed[reduce-overhead-None] 7.6325ms 6.2254ms 160.6320 Ops/s 192.5503 Ops/s $\textbf{\color{#d91a1a}-16.58\%}$
test_redq_speed[reduce-overhead-backward] 16.0153ms 13.6927ms 73.0316 Ops/s 78.0002 Ops/s $\textbf{\color{#d91a1a}-6.37\%}$
test_redq_deprec_speed[False-None] 16.5037ms 14.2391ms 70.2290 Ops/s 76.6870 Ops/s $\textbf{\color{#d91a1a}-8.42\%}$
test_redq_deprec_speed[False-backward] 22.2637ms 20.3710ms 49.0894 Ops/s 53.1171 Ops/s $\textbf{\color{#d91a1a}-7.58\%}$
test_redq_deprec_speed[True-None] 5.6427ms 4.5412ms 220.2070 Ops/s 251.9278 Ops/s $\textbf{\color{#d91a1a}-12.59\%}$
test_redq_deprec_speed[True-backward] 10.2738ms 9.6899ms 103.2003 Ops/s 110.7611 Ops/s $\textbf{\color{#d91a1a}-6.83\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.8854ms 4.3424ms 230.2866 Ops/s 250.2529 Ops/s $\textbf{\color{#d91a1a}-7.98\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.9907ms 9.4332ms 106.0091 Ops/s 114.5259 Ops/s $\textbf{\color{#d91a1a}-7.44\%}$
test_td3_speed[False-None] 9.3697ms 8.5836ms 116.5006 Ops/s 119.6816 Ops/s $\color{#d91a1a}-2.66\%$
test_td3_speed[False-backward] 11.6722ms 11.1207ms 89.9224 Ops/s 92.1206 Ops/s $\color{#d91a1a}-2.39\%$
test_td3_speed[True-None] 2.1555ms 1.8492ms 540.7658 Ops/s 537.3889 Ops/s $\color{#35bf28}+0.63\%$
test_td3_speed[True-backward] 4.2109ms 3.7726ms 265.0660 Ops/s 273.7197 Ops/s $\color{#d91a1a}-3.16\%$
test_td3_speed[reduce-overhead-None] 2.1986ms 1.8263ms 547.5455 Ops/s 529.0573 Ops/s $\color{#35bf28}+3.49\%$
test_td3_speed[reduce-overhead-backward] 3.6723ms 3.5729ms 279.8849 Ops/s 289.7517 Ops/s $\color{#d91a1a}-3.41\%$
test_cql_speed[False-None] 41.6346ms 37.9328ms 26.3624 Ops/s 26.1548 Ops/s $\color{#35bf28}+0.79\%$
test_cql_speed[False-backward] 52.5437ms 48.5195ms 20.6103 Ops/s 21.2877 Ops/s $\color{#d91a1a}-3.18\%$
test_cql_speed[True-None] 18.0932ms 16.6912ms 59.9117 Ops/s 60.9234 Ops/s $\color{#d91a1a}-1.66\%$
test_cql_speed[True-backward] 25.2264ms 24.1255ms 41.4499 Ops/s 43.1983 Ops/s $\color{#d91a1a}-4.05\%$
test_cql_speed[reduce-overhead-None] 17.7909ms 16.7441ms 59.7225 Ops/s 61.3564 Ops/s $\color{#d91a1a}-2.66\%$
test_cql_speed[reduce-overhead-backward] 29.1800ms 23.6345ms 42.3110 Ops/s 43.0014 Ops/s $\color{#d91a1a}-1.61\%$
test_a2c_speed[False-None] 8.1823ms 7.2700ms 137.5523 Ops/s 135.6215 Ops/s $\color{#35bf28}+1.42\%$
test_a2c_speed[False-backward] 15.4235ms 14.5807ms 68.5837 Ops/s 67.9036 Ops/s $\color{#35bf28}+1.00\%$
test_a2c_speed[True-None] 4.4907ms 3.7238ms 268.5455 Ops/s 263.5749 Ops/s $\color{#35bf28}+1.89\%$
test_a2c_speed[True-backward] 11.8475ms 10.3554ms 96.5678 Ops/s 96.2581 Ops/s $\color{#35bf28}+0.32\%$
test_a2c_speed[reduce-overhead-None] 4.0669ms 3.7210ms 268.7448 Ops/s 255.9747 Ops/s $\color{#35bf28}+4.99\%$
test_a2c_speed[reduce-overhead-backward] 11.1412ms 10.3717ms 96.4167 Ops/s 93.1226 Ops/s $\color{#35bf28}+3.54\%$
test_ppo_speed[False-None] 9.3905ms 7.6605ms 130.5395 Ops/s 126.3707 Ops/s $\color{#35bf28}+3.30\%$
test_ppo_speed[False-backward] 16.2676ms 15.2029ms 65.7771 Ops/s 64.7139 Ops/s $\color{#35bf28}+1.64\%$
test_ppo_speed[True-None] 5.2317ms 4.1072ms 243.4750 Ops/s 229.4529 Ops/s $\textbf{\color{#35bf28}+6.11\%}$
test_ppo_speed[True-backward] 11.2175ms 10.1149ms 98.8636 Ops/s 97.7017 Ops/s $\color{#35bf28}+1.19\%$
test_ppo_speed[reduce-overhead-None] 4.8941ms 4.1000ms 243.9029 Ops/s 235.6515 Ops/s $\color{#35bf28}+3.50\%$
test_ppo_speed[reduce-overhead-backward] 10.5361ms 10.1068ms 98.9432 Ops/s 98.5612 Ops/s $\color{#35bf28}+0.39\%$
test_reinforce_speed[False-None] 8.7105ms 6.5990ms 151.5384 Ops/s 150.6260 Ops/s $\color{#35bf28}+0.61\%$
test_reinforce_speed[False-backward] 10.1459ms 9.8396ms 101.6297 Ops/s 99.2909 Ops/s $\color{#35bf28}+2.36\%$
test_reinforce_speed[True-None] 3.7618ms 3.0697ms 325.7644 Ops/s 316.0776 Ops/s $\color{#35bf28}+3.06\%$
test_reinforce_speed[True-backward] 10.1046ms 9.5554ms 104.6527 Ops/s 109.5087 Ops/s $\color{#d91a1a}-4.43\%$
test_reinforce_speed[reduce-overhead-None] 3.5861ms 3.0678ms 325.9660 Ops/s 318.6571 Ops/s $\color{#35bf28}+2.29\%$
test_reinforce_speed[reduce-overhead-backward] 9.9492ms 9.0785ms 110.1502 Ops/s 109.4383 Ops/s $\color{#35bf28}+0.65\%$
test_iql_speed[False-None] 34.6112ms 32.4902ms 30.7785 Ops/s 30.4531 Ops/s $\color{#35bf28}+1.07\%$
test_iql_speed[False-backward] 47.2581ms 45.3821ms 22.0351 Ops/s 21.8754 Ops/s $\color{#35bf28}+0.73\%$
test_iql_speed[True-None] 12.9690ms 11.4452ms 87.3727 Ops/s 84.8015 Ops/s $\color{#35bf28}+3.03\%$
test_iql_speed[True-backward] 23.4025ms 22.3847ms 44.6734 Ops/s 44.2377 Ops/s $\color{#35bf28}+0.98\%$
test_iql_speed[reduce-overhead-None] 14.9325ms 12.0501ms 82.9869 Ops/s 86.3307 Ops/s $\color{#d91a1a}-3.87\%$
test_iql_speed[reduce-overhead-backward] 24.1192ms 23.0695ms 43.3474 Ops/s 43.6370 Ops/s $\color{#d91a1a}-0.66\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.8562ms 5.1052ms 195.8789 Ops/s 191.7361 Ops/s $\color{#35bf28}+2.16\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8300ms 0.5126ms 1.9507 KOps/s 1.8973 KOps/s $\color{#35bf28}+2.81\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8170ms 0.4904ms 2.0393 KOps/s 1.9873 KOps/s $\color{#35bf28}+2.61\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.2643ms 4.6779ms 213.7715 Ops/s 213.6242 Ops/s $\color{#35bf28}+0.07\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.1914ms 0.5089ms 1.9651 KOps/s 1.9704 KOps/s $\color{#d91a1a}-0.27\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7773ms 0.4868ms 2.0544 KOps/s 2.0584 KOps/s $\color{#d91a1a}-0.20\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.8781ms 1.6587ms 602.8819 Ops/s 597.8013 Ops/s $\color{#35bf28}+0.85\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.7829ms 1.5777ms 633.8377 Ops/s 633.3103 Ops/s $\color{#35bf28}+0.08\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.3490ms 4.8034ms 208.1872 Ops/s 207.5041 Ops/s $\color{#35bf28}+0.33\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.5592ms 0.6570ms 1.5220 KOps/s 1.5317 KOps/s $\color{#d91a1a}-0.63\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9906ms 0.6286ms 1.5909 KOps/s 1.5748 KOps/s $\color{#35bf28}+1.02\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0517ms 4.6957ms 212.9615 Ops/s 209.4051 Ops/s $\color{#35bf28}+1.70\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5440s 1.2321ms 811.6359 Ops/s 1.8719 KOps/s $\textbf{\color{#d91a1a}-56.64\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7132ms 0.4865ms 2.0553 KOps/s 2.0253 KOps/s $\color{#35bf28}+1.49\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 4.9563ms 4.6173ms 216.5747 Ops/s 211.3738 Ops/s $\color{#35bf28}+2.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8453ms 0.5025ms 1.9902 KOps/s 424.5497 Ops/s $\textbf{\color{#35bf28}+368.77\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8200ms 0.4837ms 2.0674 KOps/s 2.0391 KOps/s $\color{#35bf28}+1.39\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.1318ms 4.8388ms 206.6610 Ops/s 203.5356 Ops/s $\color{#35bf28}+1.54\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1009ms 0.6514ms 1.5352 KOps/s 1.5110 KOps/s $\color{#35bf28}+1.60\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.1297ms 0.6299ms 1.5876 KOps/s 1.5954 KOps/s $\color{#d91a1a}-0.49\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.5404ms 4.2000ms 238.0950 Ops/s 233.5645 Ops/s $\color{#35bf28}+1.94\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.8755ms 2.3509ms 425.3693 Ops/s 405.8871 Ops/s $\color{#35bf28}+4.80\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.9085ms 1.3837ms 722.7079 Ops/s 579.1994 Ops/s $\textbf{\color{#35bf28}+24.78\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4404s 13.0272ms 76.7626 Ops/s 231.0672 Ops/s $\textbf{\color{#d91a1a}-66.78\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 4.9662ms 2.3046ms 433.9084 Ops/s 426.7590 Ops/s $\color{#35bf28}+1.68\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.4302ms 1.4553ms 687.1325 Ops/s 699.7461 Ops/s $\color{#d91a1a}-1.80\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.5840ms 4.4881ms 222.8120 Ops/s 216.2267 Ops/s $\color{#35bf28}+3.05\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.7488ms 2.6197ms 381.7177 Ops/s 364.4500 Ops/s $\color{#35bf28}+4.74\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.0041ms 1.6069ms 622.3073 Ops/s 636.0446 Ops/s $\color{#d91a1a}-2.16\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 12.6778ms 12.1061ms 82.6027 Ops/s 81.4943 Ops/s $\color{#35bf28}+1.36\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.2929ms 14.8519ms 67.3316 Ops/s 68.2874 Ops/s $\color{#d91a1a}-1.40\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 22.5307ms 20.7883ms 48.1041 Ops/s 47.9319 Ops/s $\color{#35bf28}+0.36\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 17.3523ms 14.8348ms 67.4092 Ops/s 68.6350 Ops/s $\color{#d91a1a}-1.79\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.0774ms 20.7656ms 48.1565 Ops/s 48.0390 Ops/s $\color{#35bf28}+0.24\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 17.8612ms 16.0092ms 62.4641 Ops/s 62.9188 Ops/s $\color{#d91a1a}-0.72\%$

Copy link

github-actions bot commented Feb 12, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8971s 0.7984s 1.2525 Ops/s 1.2704 Ops/s $\color{#d91a1a}-1.41\%$
test_transformed 1.4925s 1.3977s 0.7155 Ops/s 0.7308 Ops/s $\color{#d91a1a}-2.09\%$
test_serial 2.4101s 2.2850s 0.4376 Ops/s 0.4466 Ops/s $\color{#d91a1a}-2.02\%$
test_parallel 1.9304s 1.8195s 0.5496 Ops/s 0.5520 Ops/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[True-True-True-True-True] 0.1850ms 41.1074μs 24.3265 KOps/s 25.9172 KOps/s $\textbf{\color{#d91a1a}-6.14\%}$
test_step_mdp_speed[True-True-True-True-False] 86.5910μs 23.4314μs 42.6777 KOps/s 43.3309 KOps/s $\color{#d91a1a}-1.51\%$
test_step_mdp_speed[True-True-True-False-True] 50.9210μs 22.4459μs 44.5517 KOps/s 45.1548 KOps/s $\color{#d91a1a}-1.34\%$
test_step_mdp_speed[True-True-True-False-False] 49.6610μs 13.0602μs 76.5688 KOps/s 77.6924 KOps/s $\color{#d91a1a}-1.45\%$
test_step_mdp_speed[True-True-False-True-True] 68.3610μs 43.0577μs 23.2247 KOps/s 23.8520 KOps/s $\color{#d91a1a}-2.63\%$
test_step_mdp_speed[True-True-False-True-False] 55.2410μs 25.8288μs 38.7165 KOps/s 39.8209 KOps/s $\color{#d91a1a}-2.77\%$
test_step_mdp_speed[True-True-False-False-True] 71.9110μs 25.1469μs 39.7664 KOps/s 40.7207 KOps/s $\color{#d91a1a}-2.34\%$
test_step_mdp_speed[True-True-False-False-False] 0.1036ms 15.2978μs 65.3690 KOps/s 65.6996 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-False-True-True-True] 0.1170ms 44.8868μs 22.2783 KOps/s 22.3367 KOps/s $\color{#d91a1a}-0.26\%$
test_step_mdp_speed[True-False-True-True-False] 0.2187ms 27.9585μs 35.7673 KOps/s 36.5797 KOps/s $\color{#d91a1a}-2.22\%$
test_step_mdp_speed[True-False-True-False-True] 0.2407ms 24.5687μs 40.7021 KOps/s 41.3276 KOps/s $\color{#d91a1a}-1.51\%$
test_step_mdp_speed[True-False-True-False-False] 0.2160ms 15.4847μs 64.5799 KOps/s 65.7926 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[True-False-False-True-True] 0.2477ms 47.0480μs 21.2549 KOps/s 21.5236 KOps/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[True-False-False-True-False] 59.9410μs 30.2842μs 33.0206 KOps/s 33.2023 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[True-False-False-False-True] 62.7120μs 27.4957μs 36.3694 KOps/s 38.0817 KOps/s $\color{#d91a1a}-4.50\%$
test_step_mdp_speed[True-False-False-False-False] 55.3510μs 17.5174μs 57.0861 KOps/s 57.5609 KOps/s $\color{#d91a1a}-0.82\%$
test_step_mdp_speed[False-True-True-True-True] 0.1157ms 44.1611μs 22.6444 KOps/s 22.2837 KOps/s $\color{#35bf28}+1.62\%$
test_step_mdp_speed[False-True-True-True-False] 70.3810μs 27.8933μs 35.8509 KOps/s 35.6867 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[False-True-True-False-True] 65.8410μs 28.5177μs 35.0659 KOps/s 35.4187 KOps/s $\color{#d91a1a}-1.00\%$
test_step_mdp_speed[False-True-True-False-False] 54.5810μs 17.0430μs 58.6752 KOps/s 59.8243 KOps/s $\color{#d91a1a}-1.92\%$
test_step_mdp_speed[False-True-False-True-True] 0.2250ms 47.0720μs 21.2441 KOps/s 21.6063 KOps/s $\color{#d91a1a}-1.68\%$
test_step_mdp_speed[False-True-False-True-False] 0.2233ms 30.3962μs 32.8988 KOps/s 33.6813 KOps/s $\color{#d91a1a}-2.32\%$
test_step_mdp_speed[False-True-False-False-True] 4.0113ms 31.2517μs 31.9982 KOps/s 32.5857 KOps/s $\color{#d91a1a}-1.80\%$
test_step_mdp_speed[False-True-False-False-False] 0.1859ms 19.6144μs 50.9830 KOps/s 52.2731 KOps/s $\color{#d91a1a}-2.47\%$
test_step_mdp_speed[False-False-True-True-True] 0.2334ms 49.5676μs 20.1745 KOps/s 20.4867 KOps/s $\color{#d91a1a}-1.52\%$
test_step_mdp_speed[False-False-True-True-False] 0.1072ms 32.1182μs 31.1350 KOps/s 30.9258 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[False-False-True-False-True] 65.2210μs 30.8153μs 32.4515 KOps/s 33.1066 KOps/s $\color{#d91a1a}-1.98\%$
test_step_mdp_speed[False-False-True-False-False] 71.5620μs 19.3481μs 51.6845 KOps/s 52.3594 KOps/s $\color{#d91a1a}-1.29\%$
test_step_mdp_speed[False-False-False-True-True] 0.1067ms 50.4538μs 19.8201 KOps/s 19.6703 KOps/s $\color{#35bf28}+0.76\%$
test_step_mdp_speed[False-False-False-True-False] 0.1297ms 33.4736μs 29.8743 KOps/s 29.1409 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[False-False-False-False-True] 57.6710μs 32.3878μs 30.8759 KOps/s 31.2787 KOps/s $\color{#d91a1a}-1.29\%$
test_step_mdp_speed[False-False-False-False-False] 47.8310μs 21.3773μs 46.7786 KOps/s 46.9480 KOps/s $\color{#d91a1a}-0.36\%$
test_values[generalized_advantage_estimate-True-True] 23.0895ms 22.6032ms 44.2415 Ops/s 44.0677 Ops/s $\color{#35bf28}+0.39\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1170s 3.2160ms 310.9464 Ops/s 340.0726 Ops/s $\textbf{\color{#d91a1a}-8.56\%}$
test_values[td0_return_estimate-False-False] 0.1020ms 73.8934μs 13.5330 KOps/s 13.4604 KOps/s $\color{#35bf28}+0.54\%$
test_values[td1_return_estimate-False-False] 51.9593ms 50.9324ms 19.6339 Ops/s 19.8015 Ops/s $\color{#d91a1a}-0.85\%$
test_values[vec_td1_return_estimate-False-False] 1.2961ms 1.0409ms 960.6994 Ops/s 957.6895 Ops/s $\color{#35bf28}+0.31\%$
test_values[td_lambda_return_estimate-True-False] 82.1219ms 81.1236ms 12.3269 Ops/s 12.5175 Ops/s $\color{#d91a1a}-1.52\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2036ms 1.0346ms 966.5788 Ops/s 967.2250 Ops/s $\color{#d91a1a}-0.07\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 23.1505ms 22.7696ms 43.9182 Ops/s 44.6668 Ops/s $\color{#d91a1a}-1.68\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0332ms 0.7143ms 1.4000 KOps/s 1.4075 KOps/s $\color{#d91a1a}-0.53\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7752ms 0.6366ms 1.5708 KOps/s 1.6078 KOps/s $\color{#d91a1a}-2.30\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6544ms 1.4530ms 688.2133 Ops/s 691.1496 Ops/s $\color{#d91a1a}-0.42\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8160ms 0.6419ms 1.5579 KOps/s 1.5669 KOps/s $\color{#d91a1a}-0.58\%$
test_dqn_speed[False-None] 7.9819ms 1.5048ms 664.5417 Ops/s 680.6125 Ops/s $\color{#d91a1a}-2.36\%$
test_dqn_speed[False-backward] 2.2806ms 2.0897ms 478.5369 Ops/s 490.6316 Ops/s $\color{#d91a1a}-2.47\%$
test_dqn_speed[True-None] 0.7233ms 0.5457ms 1.8325 KOps/s 1.7320 KOps/s $\textbf{\color{#35bf28}+5.81\%}$
test_dqn_speed[True-backward] 1.3466ms 1.2062ms 829.0360 Ops/s 878.0946 Ops/s $\textbf{\color{#d91a1a}-5.59\%}$
test_dqn_speed[reduce-overhead-None] 0.7464ms 0.5646ms 1.7712 KOps/s 1.7758 KOps/s $\color{#d91a1a}-0.26\%$
test_dqn_speed[reduce-overhead-backward] 1.0928ms 1.0488ms 953.5121 Ops/s 1.0539 KOps/s $\textbf{\color{#d91a1a}-9.53\%}$
test_ddpg_speed[False-None] 3.1493ms 2.8276ms 353.6534 Ops/s 355.3536 Ops/s $\color{#d91a1a}-0.48\%$
test_ddpg_speed[False-backward] 4.4470ms 4.1075ms 243.4575 Ops/s 250.9359 Ops/s $\color{#d91a1a}-2.98\%$
test_ddpg_speed[True-None] 1.4755ms 1.3100ms 763.3436 Ops/s 758.9703 Ops/s $\color{#35bf28}+0.58\%$
test_ddpg_speed[True-backward] 2.6537ms 2.5113ms 398.1955 Ops/s 422.5180 Ops/s $\textbf{\color{#d91a1a}-5.76\%}$
test_ddpg_speed[reduce-overhead-None] 1.5184ms 1.3211ms 756.9637 Ops/s 754.8959 Ops/s $\color{#35bf28}+0.27\%$
test_ddpg_speed[reduce-overhead-backward] 2.1480ms 1.9945ms 501.3848 Ops/s 536.4305 Ops/s $\textbf{\color{#d91a1a}-6.53\%}$
test_sac_speed[False-None] 8.3473ms 7.7669ms 128.7508 Ops/s 129.3573 Ops/s $\color{#d91a1a}-0.47\%$
test_sac_speed[False-backward] 11.2521ms 10.7713ms 92.8397 Ops/s 95.3907 Ops/s $\color{#d91a1a}-2.67\%$
test_sac_speed[True-None] 2.1061ms 1.8650ms 536.1970 Ops/s 551.8413 Ops/s $\color{#d91a1a}-2.83\%$
test_sac_speed[True-backward] 4.0688ms 3.6324ms 275.3022 Ops/s 274.7547 Ops/s $\color{#35bf28}+0.20\%$
test_sac_speed[reduce-overhead-None] 18.1631ms 10.9472ms 91.3473 Ops/s 92.0385 Ops/s $\color{#d91a1a}-0.75\%$
test_sac_speed[reduce-overhead-backward] 1.9120ms 1.7408ms 574.4600 Ops/s 615.0698 Ops/s $\textbf{\color{#d91a1a}-6.60\%}$
test_redq_speed[False-None] 7.6945ms 7.2270ms 138.3697 Ops/s 137.5628 Ops/s $\color{#35bf28}+0.59\%$
test_redq_speed[False-backward] 11.7333ms 11.1848ms 89.4072 Ops/s 91.6477 Ops/s $\color{#d91a1a}-2.44\%$
test_redq_speed[True-None] 2.5737ms 2.2594ms 442.5866 Ops/s 441.4009 Ops/s $\color{#35bf28}+0.27\%$
test_redq_speed[True-backward] 4.3902ms 4.0783ms 245.2021 Ops/s 256.4611 Ops/s $\color{#d91a1a}-4.39\%$
test_redq_speed[reduce-overhead-None] 2.4645ms 2.2895ms 436.7759 Ops/s 436.6385 Ops/s $\color{#35bf28}+0.03\%$
test_redq_speed[reduce-overhead-backward] 4.2396ms 4.0667ms 245.8996 Ops/s 244.6694 Ops/s $\color{#35bf28}+0.50\%$
test_redq_deprec_speed[False-None] 9.2835ms 8.7672ms 114.0620 Ops/s 115.7161 Ops/s $\color{#d91a1a}-1.43\%$
test_redq_deprec_speed[False-backward] 12.4512ms 11.8213ms 84.5929 Ops/s 85.7743 Ops/s $\color{#d91a1a}-1.38\%$
test_redq_deprec_speed[True-None] 2.8810ms 2.5652ms 389.8361 Ops/s 375.6426 Ops/s $\color{#35bf28}+3.78\%$
test_redq_deprec_speed[True-backward] 4.9716ms 4.3868ms 227.9562 Ops/s 228.9166 Ops/s $\color{#d91a1a}-0.42\%$
test_redq_deprec_speed[reduce-overhead-None] 2.9079ms 2.5975ms 384.9834 Ops/s 388.4564 Ops/s $\color{#d91a1a}-0.89\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.8053ms 4.3502ms 229.8771 Ops/s 228.9826 Ops/s $\color{#35bf28}+0.39\%$
test_td3_speed[False-None] 8.1688ms 7.8790ms 126.9193 Ops/s 131.1900 Ops/s $\color{#d91a1a}-3.26\%$
test_td3_speed[False-backward] 10.6830ms 10.0974ms 99.0356 Ops/s 99.8308 Ops/s $\color{#d91a1a}-0.80\%$
test_td3_speed[True-None] 1.7268ms 1.6478ms 606.8542 Ops/s 610.1365 Ops/s $\color{#d91a1a}-0.54\%$
test_td3_speed[True-backward] 3.4092ms 3.2588ms 306.8641 Ops/s 303.7151 Ops/s $\color{#35bf28}+1.04\%$
test_td3_speed[reduce-overhead-None] 73.6896ms 27.2751ms 36.6635 Ops/s 37.2519 Ops/s $\color{#d91a1a}-1.58\%$
test_td3_speed[reduce-overhead-backward] 1.3624ms 1.3047ms 766.4882 Ops/s 672.7263 Ops/s $\textbf{\color{#35bf28}+13.94\%}$
test_cql_speed[False-None] 16.6977ms 16.2330ms 61.6030 Ops/s 62.0217 Ops/s $\color{#d91a1a}-0.68\%$
test_cql_speed[False-backward] 21.7855ms 20.9447ms 47.7449 Ops/s 47.0503 Ops/s $\color{#35bf28}+1.48\%$
test_cql_speed[True-None] 5.2255ms 3.1873ms 313.7460 Ops/s 311.5049 Ops/s $\color{#35bf28}+0.72\%$
test_cql_speed[True-backward] 5.8597ms 5.5144ms 181.3450 Ops/s 182.1342 Ops/s $\color{#d91a1a}-0.43\%$
test_cql_speed[reduce-overhead-None] 20.0880ms 12.9453ms 77.2484 Ops/s 76.9440 Ops/s $\color{#35bf28}+0.40\%$
test_cql_speed[reduce-overhead-backward] 2.1173ms 1.9470ms 513.6210 Ops/s 502.2478 Ops/s $\color{#35bf28}+2.26\%$
test_a2c_speed[False-None] 3.4973ms 3.0866ms 323.9852 Ops/s 325.1870 Ops/s $\color{#d91a1a}-0.37\%$
test_a2c_speed[False-backward] 6.6455ms 6.0525ms 165.2214 Ops/s 165.6246 Ops/s $\color{#d91a1a}-0.24\%$
test_a2c_speed[True-None] 1.7446ms 1.3243ms 755.1199 Ops/s 752.0353 Ops/s $\color{#35bf28}+0.41\%$
test_a2c_speed[True-backward] 3.1272ms 2.9895ms 334.5044 Ops/s 332.7049 Ops/s $\color{#35bf28}+0.54\%$
test_a2c_speed[reduce-overhead-None] 14.6628ms 8.5589ms 116.8377 Ops/s 119.5779 Ops/s $\color{#d91a1a}-2.29\%$
test_a2c_speed[reduce-overhead-backward] 1.7242ms 1.5806ms 632.6729 Ops/s 652.2821 Ops/s $\color{#d91a1a}-3.01\%$
test_ppo_speed[False-None] 3.9502ms 3.5694ms 280.1580 Ops/s 281.5133 Ops/s $\color{#d91a1a}-0.48\%$
test_ppo_speed[False-backward] 7.1930ms 6.6930ms 149.4089 Ops/s 150.6128 Ops/s $\color{#d91a1a}-0.80\%$
test_ppo_speed[True-None] 1.7776ms 1.3951ms 716.8064 Ops/s 719.1332 Ops/s $\color{#d91a1a}-0.32\%$
test_ppo_speed[True-backward] 3.4582ms 3.1492ms 317.5389 Ops/s 310.3143 Ops/s $\color{#35bf28}+2.33\%$
test_ppo_speed[reduce-overhead-None] 1.1255ms 0.9526ms 1.0497 KOps/s 1.0528 KOps/s $\color{#d91a1a}-0.29\%$
test_ppo_speed[reduce-overhead-backward] 1.6986ms 1.5366ms 650.7858 Ops/s 653.1681 Ops/s $\color{#d91a1a}-0.36\%$
test_reinforce_speed[False-None] 2.4084ms 2.2139ms 451.6847 Ops/s 450.5449 Ops/s $\color{#35bf28}+0.25\%$
test_reinforce_speed[False-backward] 3.6196ms 3.2703ms 305.7801 Ops/s 302.8515 Ops/s $\color{#35bf28}+0.97\%$
test_reinforce_speed[True-None] 1.6667ms 1.2685ms 788.3136 Ops/s 775.1870 Ops/s $\color{#35bf28}+1.69\%$
test_reinforce_speed[True-backward] 3.1828ms 3.0058ms 332.6860 Ops/s 349.4188 Ops/s $\color{#d91a1a}-4.79\%$
test_reinforce_speed[reduce-overhead-None] 16.6973ms 9.4063ms 106.3116 Ops/s 110.7195 Ops/s $\color{#d91a1a}-3.98\%$
test_reinforce_speed[reduce-overhead-backward] 1.5301ms 1.4645ms 682.8198 Ops/s 663.1595 Ops/s $\color{#35bf28}+2.96\%$
test_iql_speed[False-None] 9.3145ms 8.9301ms 111.9812 Ops/s 110.7990 Ops/s $\color{#35bf28}+1.07\%$
test_iql_speed[False-backward] 12.8918ms 12.2970ms 81.3204 Ops/s 79.4684 Ops/s $\color{#35bf28}+2.33\%$
test_iql_speed[True-None] 2.3443ms 2.1834ms 457.9930 Ops/s 448.9340 Ops/s $\color{#35bf28}+2.02\%$
test_iql_speed[True-backward] 4.9194ms 4.6598ms 214.6027 Ops/s 210.9619 Ops/s $\color{#35bf28}+1.73\%$
test_iql_speed[reduce-overhead-None] 0.5262s 12.9899ms 76.9826 Ops/s 96.6587 Ops/s $\textbf{\color{#d91a1a}-20.36\%}$
test_iql_speed[reduce-overhead-backward] 2.0185ms 1.8577ms 538.3075 Ops/s 519.7870 Ops/s $\color{#35bf28}+3.56\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0151ms 6.3513ms 157.4492 Ops/s 154.5544 Ops/s $\color{#35bf28}+1.87\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6169ms 0.3266ms 3.0617 KOps/s 3.1581 KOps/s $\color{#d91a1a}-3.05\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7866ms 0.2687ms 3.7216 KOps/s 3.4786 KOps/s $\textbf{\color{#35bf28}+6.98\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.4619ms 6.0895ms 164.2175 Ops/s 162.3349 Ops/s $\color{#35bf28}+1.16\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.4351ms 0.3149ms 3.1758 KOps/s 2.9033 KOps/s $\textbf{\color{#35bf28}+9.39\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6436ms 0.2867ms 3.4882 KOps/s 4.1414 KOps/s $\textbf{\color{#d91a1a}-15.77\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6343ms 1.3050ms 766.2918 Ops/s 795.3482 Ops/s $\color{#d91a1a}-3.65\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4396ms 1.2223ms 818.1579 Ops/s 761.5929 Ops/s $\textbf{\color{#35bf28}+7.43\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4985ms 6.2787ms 159.2679 Ops/s 155.9849 Ops/s $\color{#35bf28}+2.10\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1735ms 0.4354ms 2.2966 KOps/s 2.2750 KOps/s $\color{#35bf28}+0.95\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6612ms 0.3932ms 2.5435 KOps/s 2.2748 KOps/s $\textbf{\color{#35bf28}+11.81\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2931ms 6.1014ms 163.8962 Ops/s 159.0760 Ops/s $\color{#35bf28}+3.03\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7924ms 0.2736ms 3.6548 KOps/s 3.3011 KOps/s $\textbf{\color{#35bf28}+10.71\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7780ms 0.3422ms 2.9223 KOps/s 3.5152 KOps/s $\textbf{\color{#d91a1a}-16.87\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 10.2542ms 6.2074ms 161.0976 Ops/s 162.0062 Ops/s $\color{#d91a1a}-0.56\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.8830ms 0.3037ms 3.2927 KOps/s 2.8400 KOps/s $\textbf{\color{#35bf28}+15.94\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4996ms 0.2633ms 3.7981 KOps/s 3.5876 KOps/s $\textbf{\color{#35bf28}+5.87\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5321ms 6.2447ms 160.1360 Ops/s 157.3694 Ops/s $\color{#35bf28}+1.76\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0672ms 0.4867ms 2.0548 KOps/s 2.3047 KOps/s $\textbf{\color{#d91a1a}-10.84\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6842ms 0.4753ms 2.1038 KOps/s 2.3482 KOps/s $\textbf{\color{#d91a1a}-10.41\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.9875ms 5.6525ms 176.9132 Ops/s 180.0599 Ops/s $\color{#d91a1a}-1.75\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.7751ms 2.2520ms 444.0490 Ops/s 433.3601 Ops/s $\color{#35bf28}+2.47\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.1531ms 1.3843ms 722.3856 Ops/s 874.3257 Ops/s $\textbf{\color{#d91a1a}-17.38\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5505s 16.4614ms 60.7482 Ops/s 182.1864 Ops/s $\textbf{\color{#d91a1a}-66.66\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.8848ms 2.0569ms 486.1641 Ops/s 470.3120 Ops/s $\color{#35bf28}+3.37\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.9697ms 0.9589ms 1.0428 KOps/s 789.7301 Ops/s $\textbf{\color{#35bf28}+32.05\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.4797ms 5.6896ms 175.7606 Ops/s 27.0543 Ops/s $\textbf{\color{#35bf28}+549.66\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.7860ms 2.2290ms 448.6329 Ops/s 433.0662 Ops/s $\color{#35bf28}+3.59\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 3.5522ms 1.3062ms 765.6013 Ops/s 684.0790 Ops/s $\textbf{\color{#35bf28}+11.92\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.2456ms 12.8498ms 77.8220 Ops/s 73.3887 Ops/s $\textbf{\color{#35bf28}+6.04\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.2343ms 16.9579ms 58.9695 Ops/s 60.1127 Ops/s $\color{#d91a1a}-1.90\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.5212ms 17.8850ms 55.9127 Ops/s 55.0044 Ops/s $\color{#35bf28}+1.65\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.0860ms 16.5952ms 60.2583 Ops/s 58.3517 Ops/s $\color{#35bf28}+3.27\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 0.4714s 26.7212ms 37.4234 Ops/s 55.7855 Ops/s $\textbf{\color{#d91a1a}-32.92\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.6968ms 18.4338ms 54.2481 Ops/s 53.9476 Ops/s $\color{#35bf28}+0.56\%$

@vmoens vmoens added the bug Something isn't working label Feb 12, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 12, 2025
ghstack-source-id: 38b90fe50eb6ad6849c1d47e507098565c08f769
Pull Request resolved: #2784
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working 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