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

[Deprecation] Remove NormalParamWrapper #2747

Merged
merged 4 commits into from
Feb 4, 2025
Merged

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 3, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2747

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

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

vmoens added a commit that referenced this pull request Feb 3, 2025
ghstack-source-id: 0a5df0abf3303211a93d5ee00144794c57e2c8c1
Pull Request resolved: #2747
@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 3, 2025
[ghstack-poisoned]
[ghstack-poisoned]
Copy link

github-actions bot commented Feb 3, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5355s 0.4481s 2.2314 Ops/s 2.2152 Ops/s $\color{#35bf28}+0.73\%$
test_transformed 1.0272s 0.9359s 1.0685 Ops/s 1.0891 Ops/s $\color{#d91a1a}-1.89\%$
test_serial 1.4608s 1.3744s 0.7276 Ops/s 0.7309 Ops/s $\color{#d91a1a}-0.45\%$
test_parallel 1.2936s 1.2060s 0.8292 Ops/s 0.8164 Ops/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[True-True-True-True-True] 0.1927ms 29.6575μs 33.7183 KOps/s 33.1010 KOps/s $\color{#35bf28}+1.87\%$
test_step_mdp_speed[True-True-True-True-False] 55.5530μs 17.4073μs 57.4471 KOps/s 56.4755 KOps/s $\color{#35bf28}+1.72\%$
test_step_mdp_speed[True-True-True-False-True] 53.4190μs 16.8144μs 59.4727 KOps/s 58.2352 KOps/s $\color{#35bf28}+2.12\%$
test_step_mdp_speed[True-True-True-False-False] 62.1960μs 9.9698μs 100.3032 KOps/s 100.7556 KOps/s $\color{#d91a1a}-0.45\%$
test_step_mdp_speed[True-True-False-True-True] 66.4740μs 32.0783μs 31.1738 KOps/s 31.2691 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-True-False-True-False] 71.5030μs 19.1769μs 52.1460 KOps/s 51.2257 KOps/s $\color{#35bf28}+1.80\%$
test_step_mdp_speed[True-True-False-False-True] 74.2980μs 18.6637μs 53.5799 KOps/s 53.2669 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[True-True-False-False-False] 43.6610μs 11.6190μs 86.0660 KOps/s 84.4691 KOps/s $\color{#35bf28}+1.89\%$
test_step_mdp_speed[True-False-True-True-True] 0.6021ms 33.7349μs 29.6429 KOps/s 29.4921 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-False-True-True-False] 62.9980μs 21.4324μs 46.6584 KOps/s 46.8634 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[True-False-True-False-True] 60.1320μs 18.7335μs 53.3804 KOps/s 53.1818 KOps/s $\color{#35bf28}+0.37\%$
test_step_mdp_speed[True-False-True-False-False] 34.6750μs 11.7729μs 84.9406 KOps/s 84.1970 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[True-False-False-True-True] 74.1880μs 35.2912μs 28.3356 KOps/s 28.2497 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[True-False-False-True-False] 64.9010μs 22.8581μs 43.7481 KOps/s 43.0656 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-False-False-False-True] 50.2240μs 20.2938μs 49.2762 KOps/s 48.6466 KOps/s $\color{#35bf28}+1.29\%$
test_step_mdp_speed[True-False-False-False-False] 52.9990μs 13.3331μs 75.0013 KOps/s 73.2117 KOps/s $\color{#35bf28}+2.44\%$
test_step_mdp_speed[False-True-True-True-True] 94.4260μs 33.6527μs 29.7153 KOps/s 29.4493 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[False-True-True-True-False] 52.0970μs 21.3597μs 46.8171 KOps/s 46.7330 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-True-True-False-True] 83.9860μs 21.3664μs 46.8024 KOps/s 46.1888 KOps/s $\color{#35bf28}+1.33\%$
test_step_mdp_speed[False-True-True-False-False] 41.8770μs 13.0319μs 76.7347 KOps/s 75.7952 KOps/s $\color{#35bf28}+1.24\%$
test_step_mdp_speed[False-True-False-True-True] 90.9490μs 35.2593μs 28.3613 KOps/s 28.0925 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[False-True-False-True-False] 54.8010μs 23.1391μs 43.2170 KOps/s 43.4064 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[False-True-False-False-True] 2.6624ms 22.9106μs 43.6479 KOps/s 43.1506 KOps/s $\color{#35bf28}+1.15\%$
test_step_mdp_speed[False-True-False-False-False] 59.3910μs 14.6579μs 68.2227 KOps/s 67.3160 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[False-False-True-True-True] 71.6430μs 36.9483μs 27.0648 KOps/s 26.7965 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[False-False-True-True-False] 0.5853ms 24.9371μs 40.1009 KOps/s 40.3053 KOps/s $\color{#d91a1a}-0.51\%$
test_step_mdp_speed[False-False-True-False-True] 75.3500μs 23.3021μs 42.9145 KOps/s 43.6721 KOps/s $\color{#d91a1a}-1.73\%$
test_step_mdp_speed[False-False-True-False-False] 59.5980μs 14.5914μs 68.5335 KOps/s 67.8649 KOps/s $\color{#35bf28}+0.99\%$
test_step_mdp_speed[False-False-False-True-True] 85.3390μs 38.2969μs 26.1118 KOps/s 25.6451 KOps/s $\color{#35bf28}+1.82\%$
test_step_mdp_speed[False-False-False-True-False] 60.7230μs 26.3034μs 38.0179 KOps/s 37.8451 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[False-False-False-False-True] 73.7390μs 24.0790μs 41.5299 KOps/s 40.9202 KOps/s $\color{#35bf28}+1.49\%$
test_step_mdp_speed[False-False-False-False-False] 37.6700μs 16.2224μs 61.6430 KOps/s 61.1576 KOps/s $\color{#35bf28}+0.79\%$
test_values[generalized_advantage_estimate-True-True] 10.1243ms 9.7448ms 102.6185 Ops/s 101.7759 Ops/s $\color{#35bf28}+0.83\%$
test_values[vec_generalized_advantage_estimate-True-True] 28.0264ms 26.1167ms 38.2896 Ops/s 41.0917 Ops/s $\textbf{\color{#d91a1a}-6.82\%}$
test_values[td0_return_estimate-False-False] 0.2756ms 0.1785ms 5.6036 KOps/s 5.6175 KOps/s $\color{#d91a1a}-0.25\%$
test_values[td1_return_estimate-False-False] 24.6575ms 24.1428ms 41.4202 Ops/s 41.4345 Ops/s $\color{#d91a1a}-0.03\%$
test_values[vec_td1_return_estimate-False-False] 28.6971ms 26.2885ms 38.0394 Ops/s 40.8091 Ops/s $\textbf{\color{#d91a1a}-6.79\%}$
test_values[td_lambda_return_estimate-True-False] 38.2080ms 34.9429ms 28.6182 Ops/s 28.6188 Ops/s $-0.00\%$
test_values[vec_td_lambda_return_estimate-True-False] 28.7144ms 26.3255ms 37.9859 Ops/s 40.7770 Ops/s $\textbf{\color{#d91a1a}-6.84\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.6536ms 8.4811ms 117.9086 Ops/s 117.2979 Ops/s $\color{#35bf28}+0.52\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3744ms 1.9697ms 507.6990 Ops/s 551.2792 Ops/s $\textbf{\color{#d91a1a}-7.91\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4326ms 0.3584ms 2.7905 KOps/s 2.7561 KOps/s $\color{#35bf28}+1.25\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 48.5860ms 46.3700ms 21.5657 Ops/s 24.9603 Ops/s $\textbf{\color{#d91a1a}-13.60\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.0222ms 3.4160ms 292.7425 Ops/s 292.3739 Ops/s $\color{#35bf28}+0.13\%$
test_dqn_speed[False-None] 6.1218ms 1.3977ms 715.4701 Ops/s 711.9999 Ops/s $\color{#35bf28}+0.49\%$
test_dqn_speed[False-backward] 1.9384ms 1.8588ms 537.9912 Ops/s 532.1362 Ops/s $\color{#35bf28}+1.10\%$
test_dqn_speed[True-None] 0.5838ms 0.4695ms 2.1298 KOps/s 1.9793 KOps/s $\textbf{\color{#35bf28}+7.61\%}$
test_dqn_speed[True-backward] 0.9395ms 0.8861ms 1.1285 KOps/s 1.0885 KOps/s $\color{#35bf28}+3.68\%$
test_dqn_speed[reduce-overhead-None] 0.7541ms 0.4782ms 2.0913 KOps/s 2.0601 KOps/s $\color{#35bf28}+1.52\%$
test_dqn_speed[reduce-overhead-backward] 0.9670ms 0.8986ms 1.1128 KOps/s 1.0827 KOps/s $\color{#35bf28}+2.79\%$
test_ddpg_speed[False-None] 3.6893ms 2.8557ms 350.1782 Ops/s 348.7523 Ops/s $\color{#35bf28}+0.41\%$
test_ddpg_speed[False-backward] 4.4911ms 3.9757ms 251.5285 Ops/s 252.8391 Ops/s $\color{#d91a1a}-0.52\%$
test_ddpg_speed[True-None] 1.6711ms 1.2052ms 829.7070 Ops/s 824.3446 Ops/s $\color{#35bf28}+0.65\%$
test_ddpg_speed[True-backward] 2.1217ms 2.0831ms 480.0484 Ops/s 469.5439 Ops/s $\color{#35bf28}+2.24\%$
test_ddpg_speed[reduce-overhead-None] 1.8014ms 1.2093ms 826.9450 Ops/s 817.0189 Ops/s $\color{#35bf28}+1.21\%$
test_ddpg_speed[reduce-overhead-backward] 2.1188ms 2.0645ms 484.3686 Ops/s 471.8841 Ops/s $\color{#35bf28}+2.65\%$
test_sac_speed[False-None] 8.3594ms 7.9338ms 126.0423 Ops/s 124.0619 Ops/s $\color{#35bf28}+1.60\%$
test_sac_speed[False-backward] 13.1906ms 10.6274ms 94.0968 Ops/s 89.5766 Ops/s $\textbf{\color{#35bf28}+5.05\%}$
test_sac_speed[True-None] 2.3770ms 2.0495ms 487.9347 Ops/s 474.4367 Ops/s $\color{#35bf28}+2.85\%$
test_sac_speed[True-backward] 4.0952ms 3.7072ms 269.7490 Ops/s 268.2845 Ops/s $\color{#35bf28}+0.55\%$
test_sac_speed[reduce-overhead-None] 2.6096ms 2.0454ms 488.8995 Ops/s 477.8786 Ops/s $\color{#35bf28}+2.31\%$
test_sac_speed[reduce-overhead-backward] 3.8530ms 3.7114ms 269.4428 Ops/s 262.9278 Ops/s $\color{#35bf28}+2.48\%$
test_redq_speed[False-None] 19.6760ms 13.5363ms 73.8755 Ops/s 78.2265 Ops/s $\textbf{\color{#d91a1a}-5.56\%}$
test_redq_speed[False-backward] 23.2222ms 22.3887ms 44.6654 Ops/s 44.4273 Ops/s $\color{#35bf28}+0.54\%$
test_redq_speed[True-None] 5.9136ms 5.0004ms 199.9839 Ops/s 198.3689 Ops/s $\color{#35bf28}+0.81\%$
test_redq_speed[True-backward] 13.5734ms 12.7092ms 78.6833 Ops/s 80.1551 Ops/s $\color{#d91a1a}-1.84\%$
test_redq_speed[reduce-overhead-None] 6.4205ms 5.1907ms 192.6510 Ops/s 205.0519 Ops/s $\textbf{\color{#d91a1a}-6.05\%}$
test_redq_speed[reduce-overhead-backward] 13.1572ms 12.4917ms 80.0532 Ops/s 76.1877 Ops/s $\textbf{\color{#35bf28}+5.07\%}$
test_redq_deprec_speed[False-None] 14.3897ms 12.8454ms 77.8487 Ops/s 76.6193 Ops/s $\color{#35bf28}+1.60\%$
test_redq_deprec_speed[False-backward] 19.1668ms 18.6625ms 53.5834 Ops/s 53.8743 Ops/s $\color{#d91a1a}-0.54\%$
test_redq_deprec_speed[True-None] 4.2228ms 3.8222ms 261.6319 Ops/s 251.7105 Ops/s $\color{#35bf28}+3.94\%$
test_redq_deprec_speed[True-backward] 9.9066ms 8.3853ms 119.2564 Ops/s 121.3521 Ops/s $\color{#d91a1a}-1.73\%$
test_redq_deprec_speed[reduce-overhead-None] 5.4278ms 4.1420ms 241.4306 Ops/s 260.5389 Ops/s $\textbf{\color{#d91a1a}-7.33\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.2647ms 8.5285ms 117.2540 Ops/s 121.4215 Ops/s $\color{#d91a1a}-3.43\%$
test_td3_speed[False-None] 8.6175ms 7.9749ms 125.3935 Ops/s 123.7370 Ops/s $\color{#35bf28}+1.34\%$
test_td3_speed[False-backward] 10.5947ms 10.3020ms 97.0686 Ops/s 94.7874 Ops/s $\color{#35bf28}+2.41\%$
test_td3_speed[True-None] 2.0266ms 1.7864ms 559.7770 Ops/s 563.5785 Ops/s $\color{#d91a1a}-0.67\%$
test_td3_speed[True-backward] 3.4305ms 3.3608ms 297.5451 Ops/s 295.3712 Ops/s $\color{#35bf28}+0.74\%$
test_td3_speed[reduce-overhead-None] 1.9845ms 1.7892ms 558.9032 Ops/s 557.6861 Ops/s $\color{#35bf28}+0.22\%$
test_td3_speed[reduce-overhead-backward] 3.4324ms 3.3687ms 296.8527 Ops/s 295.1773 Ops/s $\color{#35bf28}+0.57\%$
test_cql_speed[False-None] 37.1700ms 35.9894ms 27.7860 Ops/s 27.2419 Ops/s $\color{#35bf28}+2.00\%$
test_cql_speed[False-backward] 49.9145ms 46.6661ms 21.4288 Ops/s 20.8699 Ops/s $\color{#35bf28}+2.68\%$
test_cql_speed[True-None] 17.2679ms 16.1348ms 61.9778 Ops/s 62.3423 Ops/s $\color{#d91a1a}-0.58\%$
test_cql_speed[True-backward] 24.3265ms 23.0832ms 43.3216 Ops/s 42.6375 Ops/s $\color{#35bf28}+1.60\%$
test_cql_speed[reduce-overhead-None] 17.4154ms 16.0658ms 62.2439 Ops/s 62.2222 Ops/s $\color{#35bf28}+0.03\%$
test_cql_speed[reduce-overhead-backward] 24.6624ms 23.3449ms 42.8358 Ops/s 42.3075 Ops/s $\color{#35bf28}+1.25\%$
test_a2c_speed[False-None] 8.1135ms 7.1420ms 140.0173 Ops/s 138.0105 Ops/s $\color{#35bf28}+1.45\%$
test_a2c_speed[False-backward] 16.2539ms 14.4962ms 68.9836 Ops/s 69.6463 Ops/s $\color{#d91a1a}-0.95\%$
test_a2c_speed[True-None] 4.3818ms 3.6921ms 270.8490 Ops/s 264.1580 Ops/s $\color{#35bf28}+2.53\%$
test_a2c_speed[True-backward] 10.6735ms 10.1410ms 98.6092 Ops/s 97.6017 Ops/s $\color{#35bf28}+1.03\%$
test_a2c_speed[reduce-overhead-None] 4.0747ms 3.6912ms 270.9136 Ops/s 264.1683 Ops/s $\color{#35bf28}+2.55\%$
test_a2c_speed[reduce-overhead-backward] 11.4484ms 10.5059ms 95.1844 Ops/s 97.2244 Ops/s $\color{#d91a1a}-2.10\%$
test_ppo_speed[False-None] 9.0241ms 7.5752ms 132.0088 Ops/s 132.8788 Ops/s $\color{#d91a1a}-0.65\%$
test_ppo_speed[False-backward] 16.6229ms 15.0093ms 66.6253 Ops/s 67.2621 Ops/s $\color{#d91a1a}-0.95\%$
test_ppo_speed[True-None] 4.7563ms 4.0865ms 244.7056 Ops/s 243.7397 Ops/s $\color{#35bf28}+0.40\%$
test_ppo_speed[True-backward] 10.2042ms 9.9563ms 100.4385 Ops/s 100.0620 Ops/s $\color{#35bf28}+0.38\%$
test_ppo_speed[reduce-overhead-None] 4.8256ms 4.0603ms 246.2859 Ops/s 243.9419 Ops/s $\color{#35bf28}+0.96\%$
test_ppo_speed[reduce-overhead-backward] 10.5314ms 9.9408ms 100.5956 Ops/s 98.7559 Ops/s $\color{#35bf28}+1.86\%$
test_reinforce_speed[False-None] 7.4226ms 6.5174ms 153.4343 Ops/s 151.2649 Ops/s $\color{#35bf28}+1.43\%$
test_reinforce_speed[False-backward] 11.3322ms 10.1100ms 98.9119 Ops/s 101.3302 Ops/s $\color{#d91a1a}-2.39\%$
test_reinforce_speed[True-None] 3.7977ms 3.0480ms 328.0862 Ops/s 325.7369 Ops/s $\color{#35bf28}+0.72\%$
test_reinforce_speed[True-backward] 9.2216ms 8.9088ms 112.2487 Ops/s 110.5121 Ops/s $\color{#35bf28}+1.57\%$
test_reinforce_speed[reduce-overhead-None] 3.6827ms 3.0356ms 329.4263 Ops/s 326.0456 Ops/s $\color{#35bf28}+1.04\%$
test_reinforce_speed[reduce-overhead-backward] 9.9809ms 8.9689ms 111.4963 Ops/s 111.6289 Ops/s $\color{#d91a1a}-0.12\%$
test_iql_speed[False-None] 34.0966ms 32.3846ms 30.8788 Ops/s 30.0385 Ops/s $\color{#35bf28}+2.80\%$
test_iql_speed[False-backward] 51.3995ms 45.3349ms 22.0581 Ops/s 21.7585 Ops/s $\color{#35bf28}+1.38\%$
test_iql_speed[True-None] 12.1167ms 11.3124ms 88.3986 Ops/s 87.8497 Ops/s $\color{#35bf28}+0.62\%$
test_iql_speed[True-backward] 23.5053ms 21.8517ms 45.7629 Ops/s 44.8721 Ops/s $\color{#35bf28}+1.99\%$
test_iql_speed[reduce-overhead-None] 12.9922ms 11.2807ms 88.6473 Ops/s 87.7811 Ops/s $\color{#35bf28}+0.99\%$
test_iql_speed[reduce-overhead-backward] 23.8591ms 22.6437ms 44.1623 Ops/s 44.7261 Ops/s $\color{#d91a1a}-1.26\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0624ms 5.1640ms 193.6488 Ops/s 201.2741 Ops/s $\color{#d91a1a}-3.79\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9372ms 0.5339ms 1.8730 KOps/s 1.9556 KOps/s $\color{#d91a1a}-4.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8132ms 0.5110ms 1.9571 KOps/s 2.0094 KOps/s $\color{#d91a1a}-2.60\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8234ms 4.9016ms 204.0163 Ops/s 218.0262 Ops/s $\textbf{\color{#d91a1a}-6.43\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9176ms 0.5011ms 1.9956 KOps/s 1.9467 KOps/s $\color{#35bf28}+2.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7605ms 0.4848ms 2.0628 KOps/s 2.0650 KOps/s $\color{#d91a1a}-0.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.4226ms 1.6399ms 609.7849 Ops/s 607.6606 Ops/s $\color{#35bf28}+0.35\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.0621ms 1.5438ms 647.7435 Ops/s 636.9832 Ops/s $\color{#35bf28}+1.69\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.9380ms 4.9579ms 201.7001 Ops/s 206.2355 Ops/s $\color{#d91a1a}-2.20\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1745ms 0.6461ms 1.5478 KOps/s 1.5332 KOps/s $\color{#35bf28}+0.95\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9062ms 0.6199ms 1.6131 KOps/s 1.6059 KOps/s $\color{#35bf28}+0.45\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.2072ms 4.6473ms 215.1776 Ops/s 213.0358 Ops/s $\color{#35bf28}+1.01\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.1601ms 0.5159ms 1.9383 KOps/s 1.9001 KOps/s $\color{#35bf28}+2.01\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7240ms 0.4867ms 2.0549 KOps/s 2.0210 KOps/s $\color{#35bf28}+1.68\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.6889ms 4.7056ms 212.5141 Ops/s 220.4675 Ops/s $\color{#d91a1a}-3.61\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7868ms 0.5050ms 1.9802 KOps/s 1.9927 KOps/s $\color{#d91a1a}-0.63\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7686ms 0.4867ms 2.0547 KOps/s 2.0223 KOps/s $\color{#35bf28}+1.60\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.3859ms 4.7598ms 210.0928 Ops/s 208.6027 Ops/s $\color{#35bf28}+0.71\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.4592s 1.3423ms 744.9885 Ops/s 1.5429 KOps/s $\textbf{\color{#d91a1a}-51.72\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8543ms 0.6168ms 1.6214 KOps/s 1.5723 KOps/s $\color{#35bf28}+3.12\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.5863ms 4.2065ms 237.7284 Ops/s 249.9012 Ops/s $\color{#d91a1a}-4.87\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.9997ms 2.2956ms 435.6098 Ops/s 449.7580 Ops/s $\color{#d91a1a}-3.15\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.5323ms 1.3269ms 753.6393 Ops/s 767.5658 Ops/s $\color{#d91a1a}-1.81\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.8484ms 4.2362ms 236.0626 Ops/s 32.7721 Ops/s $\textbf{\color{#35bf28}+620.32\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.4166s 10.5920ms 94.4107 Ops/s 415.0931 Ops/s $\textbf{\color{#d91a1a}-77.26\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.7149ms 1.3102ms 763.2325 Ops/s 727.4409 Ops/s $\color{#35bf28}+4.92\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 5.9827ms 4.3995ms 227.2991 Ops/s 223.5141 Ops/s $\color{#35bf28}+1.69\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.5873ms 2.4344ms 410.7872 Ops/s 401.6109 Ops/s $\color{#35bf28}+2.28\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 4.9622ms 1.4457ms 691.7035 Ops/s 707.7679 Ops/s $\color{#d91a1a}-2.27\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 14.0284ms 11.2660ms 88.7626 Ops/s 83.0767 Ops/s $\textbf{\color{#35bf28}+6.84\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 15.8664ms 14.2193ms 70.3268 Ops/s 69.8387 Ops/s $\color{#35bf28}+0.70\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 21.2189ms 20.0017ms 49.9958 Ops/s 47.3394 Ops/s $\textbf{\color{#35bf28}+5.61\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.4058ms 14.3921ms 69.4827 Ops/s 68.9171 Ops/s $\color{#35bf28}+0.82\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.3080ms 19.9822ms 50.0445 Ops/s 48.2673 Ops/s $\color{#35bf28}+3.68\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 16.5730ms 15.6582ms 63.8644 Ops/s 62.6122 Ops/s $\color{#35bf28}+2.00\%$

Copy link

github-actions bot commented Feb 3, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}14$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8350s 0.7483s 1.3363 Ops/s 1.3333 Ops/s $\color{#35bf28}+0.23\%$
test_transformed 1.4460s 1.3580s 0.7364 Ops/s 0.7558 Ops/s $\color{#d91a1a}-2.56\%$
test_serial 2.1822s 2.1735s 0.4601 Ops/s 0.4559 Ops/s $\color{#35bf28}+0.92\%$
test_parallel 1.8737s 1.8387s 0.5439 Ops/s 0.5358 Ops/s $\color{#35bf28}+1.51\%$
test_step_mdp_speed[True-True-True-True-True] 0.2124ms 40.5328μs 24.6714 KOps/s 24.7910 KOps/s $\color{#d91a1a}-0.48\%$
test_step_mdp_speed[True-True-True-True-False] 0.2070ms 24.4318μs 40.9302 KOps/s 42.5161 KOps/s $\color{#d91a1a}-3.73\%$
test_step_mdp_speed[True-True-True-False-True] 0.2301ms 23.7198μs 42.1590 KOps/s 44.4041 KOps/s $\textbf{\color{#d91a1a}-5.06\%}$
test_step_mdp_speed[True-True-True-False-False] 0.2024ms 13.7847μs 72.5440 KOps/s 75.9043 KOps/s $\color{#d91a1a}-4.43\%$
test_step_mdp_speed[True-True-False-True-True] 0.2219ms 43.8466μs 22.8068 KOps/s 23.2877 KOps/s $\color{#d91a1a}-2.07\%$
test_step_mdp_speed[True-True-False-True-False] 0.1131ms 26.5964μs 37.5991 KOps/s 38.7083 KOps/s $\color{#d91a1a}-2.87\%$
test_step_mdp_speed[True-True-False-False-True] 0.1126ms 26.0758μs 38.3498 KOps/s 40.2491 KOps/s $\color{#d91a1a}-4.72\%$
test_step_mdp_speed[True-True-False-False-False] 45.8620μs 16.1356μs 61.9748 KOps/s 64.0486 KOps/s $\color{#d91a1a}-3.24\%$
test_step_mdp_speed[True-False-True-True-True] 0.1484ms 45.5082μs 21.9740 KOps/s 21.8514 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-False-True-True-False] 0.1134ms 28.5261μs 35.0556 KOps/s 35.2219 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-False-True-False-True] 0.1241ms 25.8082μs 38.7474 KOps/s 40.6306 KOps/s $\color{#d91a1a}-4.63\%$
test_step_mdp_speed[True-False-True-False-False] 48.1830μs 15.9901μs 62.5388 KOps/s 64.7606 KOps/s $\color{#d91a1a}-3.43\%$
test_step_mdp_speed[True-False-False-True-True] 0.1603ms 48.3818μs 20.6689 KOps/s 20.9519 KOps/s $\color{#d91a1a}-1.35\%$
test_step_mdp_speed[True-False-False-True-False] 0.1244ms 31.3851μs 31.8622 KOps/s 32.5481 KOps/s $\color{#d91a1a}-2.11\%$
test_step_mdp_speed[True-False-False-False-True] 0.1623ms 27.5728μs 36.2676 KOps/s 36.8067 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[True-False-False-False-False] 42.7920μs 17.7344μs 56.3876 KOps/s 56.4672 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[False-True-True-True-True] 0.1669ms 45.4748μs 21.9902 KOps/s 22.2941 KOps/s $\color{#d91a1a}-1.36\%$
test_step_mdp_speed[False-True-True-True-False] 0.1639ms 29.1429μs 34.3137 KOps/s 35.8098 KOps/s $\color{#d91a1a}-4.18\%$
test_step_mdp_speed[False-True-True-False-True] 2.5080ms 29.8014μs 33.5554 KOps/s 34.7965 KOps/s $\color{#d91a1a}-3.57\%$
test_step_mdp_speed[False-True-True-False-False] 44.6120μs 17.1434μs 58.3315 KOps/s 58.0367 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[False-True-False-True-True] 0.2050ms 47.8676μs 20.8909 KOps/s 21.2440 KOps/s $\color{#d91a1a}-1.66\%$
test_step_mdp_speed[False-True-False-True-False] 0.1885ms 30.8198μs 32.4467 KOps/s 32.7373 KOps/s $\color{#d91a1a}-0.89\%$
test_step_mdp_speed[False-True-False-False-True] 0.2174ms 31.5547μs 31.6910 KOps/s 31.9179 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[False-True-False-False-False] 0.1884ms 20.2254μs 49.4428 KOps/s 50.8670 KOps/s $\color{#d91a1a}-2.80\%$
test_step_mdp_speed[False-False-True-True-True] 0.2621ms 50.6813μs 19.7311 KOps/s 19.8567 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[False-False-True-True-False] 64.0740μs 33.5917μs 29.7693 KOps/s 30.0484 KOps/s $\color{#d91a1a}-0.93\%$
test_step_mdp_speed[False-False-True-False-True] 61.8940μs 30.9295μs 32.3316 KOps/s 32.5308 KOps/s $\color{#d91a1a}-0.61\%$
test_step_mdp_speed[False-False-True-False-False] 0.1583ms 20.3996μs 49.0206 KOps/s 50.9563 KOps/s $\color{#d91a1a}-3.80\%$
test_step_mdp_speed[False-False-False-True-True] 0.1369ms 52.5034μs 19.0464 KOps/s 19.1611 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[False-False-False-True-False] 0.1330ms 35.6731μs 28.0323 KOps/s 28.5358 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[False-False-False-False-True] 60.6440μs 33.0222μs 30.2826 KOps/s 30.2613 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[False-False-False-False-False] 0.1076ms 21.9017μs 45.6586 KOps/s 45.9145 KOps/s $\color{#d91a1a}-0.56\%$
test_values[generalized_advantage_estimate-True-True] 27.3982ms 25.8031ms 38.7551 Ops/s 39.2835 Ops/s $\color{#d91a1a}-1.35\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1017s 2.9424ms 339.8630 Ops/s 338.8396 Ops/s $\color{#35bf28}+0.30\%$
test_values[td0_return_estimate-False-False] 0.1068ms 81.1823μs 12.3180 KOps/s 12.3350 KOps/s $\color{#d91a1a}-0.14\%$
test_values[td1_return_estimate-False-False] 60.9223ms 57.9701ms 17.2503 Ops/s 17.7486 Ops/s $\color{#d91a1a}-2.81\%$
test_values[vec_td1_return_estimate-False-False] 1.4109ms 1.1043ms 905.5742 Ops/s 915.8961 Ops/s $\color{#d91a1a}-1.13\%$
test_values[td_lambda_return_estimate-True-False] 95.4859ms 91.9897ms 10.8708 Ops/s 11.1903 Ops/s $\color{#d91a1a}-2.85\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4344ms 1.1111ms 899.9827 Ops/s 920.7215 Ops/s $\color{#d91a1a}-2.25\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 27.6529ms 26.3048ms 38.0159 Ops/s 39.6249 Ops/s $\color{#d91a1a}-4.06\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0354ms 0.7622ms 1.3119 KOps/s 1.3248 KOps/s $\color{#d91a1a}-0.97\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8933ms 0.6968ms 1.4351 KOps/s 1.4843 KOps/s $\color{#d91a1a}-3.31\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6837ms 1.4962ms 668.3796 Ops/s 670.2554 Ops/s $\color{#d91a1a}-0.28\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8799ms 0.6936ms 1.4417 KOps/s 1.4493 KOps/s $\color{#d91a1a}-0.52\%$
test_dqn_speed[False-None] 1.7395ms 1.5438ms 647.7479 Ops/s 659.8474 Ops/s $\color{#d91a1a}-1.83\%$
test_dqn_speed[False-backward] 2.4466ms 2.1696ms 460.9073 Ops/s 466.5831 Ops/s $\color{#d91a1a}-1.22\%$
test_dqn_speed[True-None] 0.7048ms 0.5567ms 1.7962 KOps/s 1.7731 KOps/s $\color{#35bf28}+1.30\%$
test_dqn_speed[True-backward] 1.4091ms 1.2339ms 810.4538 Ops/s 868.2049 Ops/s $\textbf{\color{#d91a1a}-6.65\%}$
test_dqn_speed[reduce-overhead-None] 0.7866ms 0.5992ms 1.6689 KOps/s 1.7028 KOps/s $\color{#d91a1a}-1.99\%$
test_dqn_speed[reduce-overhead-backward] 1.1126ms 1.0675ms 936.8020 Ops/s 1.0386 KOps/s $\textbf{\color{#d91a1a}-9.80\%}$
test_ddpg_speed[False-None] 3.2286ms 2.9295ms 341.3514 Ops/s 341.6484 Ops/s $\color{#d91a1a}-0.09\%$
test_ddpg_speed[False-backward] 5.0227ms 4.3333ms 230.7684 Ops/s 242.3799 Ops/s $\color{#d91a1a}-4.79\%$
test_ddpg_speed[True-None] 1.6013ms 1.3845ms 722.2632 Ops/s 741.0312 Ops/s $\color{#d91a1a}-2.53\%$
test_ddpg_speed[True-backward] 2.9350ms 2.6588ms 376.1152 Ops/s 388.7188 Ops/s $\color{#d91a1a}-3.24\%$
test_ddpg_speed[reduce-overhead-None] 1.6307ms 1.3654ms 732.3877 Ops/s 734.5561 Ops/s $\color{#d91a1a}-0.30\%$
test_ddpg_speed[reduce-overhead-backward] 2.3866ms 2.0412ms 489.8979 Ops/s 524.5394 Ops/s $\textbf{\color{#d91a1a}-6.60\%}$
test_sac_speed[False-None] 8.8743ms 8.3122ms 120.3056 Ops/s 123.2058 Ops/s $\color{#d91a1a}-2.35\%$
test_sac_speed[False-backward] 11.8710ms 11.4171ms 87.5881 Ops/s 90.6111 Ops/s $\color{#d91a1a}-3.34\%$
test_sac_speed[True-None] 2.1503ms 1.8513ms 540.1737 Ops/s 538.0228 Ops/s $\color{#35bf28}+0.40\%$
test_sac_speed[True-backward] 3.8097ms 3.5803ms 279.3065 Ops/s 277.3929 Ops/s $\color{#35bf28}+0.69\%$
test_sac_speed[reduce-overhead-None] 21.5084ms 12.2024ms 81.9513 Ops/s 80.6502 Ops/s $\color{#35bf28}+1.61\%$
test_sac_speed[reduce-overhead-backward] 1.8186ms 1.6472ms 607.0936 Ops/s 547.0979 Ops/s $\textbf{\color{#35bf28}+10.97\%}$
test_redq_speed[False-None] 8.0622ms 7.6188ms 131.2534 Ops/s 131.5562 Ops/s $\color{#d91a1a}-0.23\%$
test_redq_speed[False-backward] 11.9275ms 11.4384ms 87.4246 Ops/s 84.4091 Ops/s $\color{#35bf28}+3.57\%$
test_redq_speed[True-None] 2.6280ms 2.3274ms 429.6715 Ops/s 427.7573 Ops/s $\color{#35bf28}+0.45\%$
test_redq_speed[True-backward] 4.4929ms 4.0237ms 248.5270 Ops/s 245.6086 Ops/s $\color{#35bf28}+1.19\%$
test_redq_speed[reduce-overhead-None] 2.7198ms 2.3380ms 427.7106 Ops/s 419.5833 Ops/s $\color{#35bf28}+1.94\%$
test_redq_speed[reduce-overhead-backward] 4.5557ms 4.0345ms 247.8597 Ops/s 242.8181 Ops/s $\color{#35bf28}+2.08\%$
test_redq_deprec_speed[False-None] 9.6785ms 9.2332ms 108.3049 Ops/s 108.2012 Ops/s $\color{#35bf28}+0.10\%$
test_redq_deprec_speed[False-backward] 12.9962ms 12.2916ms 81.3564 Ops/s 81.8895 Ops/s $\color{#d91a1a}-0.65\%$
test_redq_deprec_speed[True-None] 3.1051ms 2.6527ms 376.9787 Ops/s 371.0738 Ops/s $\color{#35bf28}+1.59\%$
test_redq_deprec_speed[True-backward] 4.9605ms 4.5117ms 221.6452 Ops/s 221.5441 Ops/s $\color{#35bf28}+0.05\%$
test_redq_deprec_speed[reduce-overhead-None] 3.0449ms 2.6497ms 377.3997 Ops/s 369.9889 Ops/s $\color{#35bf28}+2.00\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.9501ms 4.5472ms 219.9149 Ops/s 228.3036 Ops/s $\color{#d91a1a}-3.67\%$
test_td3_speed[False-None] 8.5512ms 8.2268ms 121.5533 Ops/s 122.5613 Ops/s $\color{#d91a1a}-0.82\%$
test_td3_speed[False-backward] 11.4251ms 10.8545ms 92.1281 Ops/s 96.2899 Ops/s $\color{#d91a1a}-4.32\%$
test_td3_speed[True-None] 1.7720ms 1.7232ms 580.3303 Ops/s 584.2473 Ops/s $\color{#d91a1a}-0.67\%$
test_td3_speed[True-backward] 3.6055ms 3.2196ms 310.5949 Ops/s 307.3005 Ops/s $\color{#35bf28}+1.07\%$
test_td3_speed[reduce-overhead-None] 54.2006ms 26.4511ms 37.8057 Ops/s 36.2725 Ops/s $\color{#35bf28}+4.23\%$
test_td3_speed[reduce-overhead-backward] 1.9118ms 1.5191ms 658.2870 Ops/s 710.7823 Ops/s $\textbf{\color{#d91a1a}-7.39\%}$
test_cql_speed[False-None] 17.4396ms 17.0358ms 58.6998 Ops/s 58.9818 Ops/s $\color{#d91a1a}-0.48\%$
test_cql_speed[False-backward] 23.2713ms 22.5075ms 44.4296 Ops/s 45.3242 Ops/s $\color{#d91a1a}-1.97\%$
test_cql_speed[True-None] 3.5733ms 3.2649ms 306.2870 Ops/s 304.2593 Ops/s $\color{#35bf28}+0.67\%$
test_cql_speed[True-backward] 6.2061ms 5.7406ms 174.1986 Ops/s 175.4758 Ops/s $\color{#d91a1a}-0.73\%$
test_cql_speed[reduce-overhead-None] 21.0081ms 13.1942ms 75.7907 Ops/s 57.5266 Ops/s $\textbf{\color{#35bf28}+31.75\%}$
test_cql_speed[reduce-overhead-backward] 2.1797ms 1.9951ms 501.2258 Ops/s 534.0322 Ops/s $\textbf{\color{#d91a1a}-6.14\%}$
test_a2c_speed[False-None] 3.4574ms 3.2491ms 307.7748 Ops/s 308.4906 Ops/s $\color{#d91a1a}-0.23\%$
test_a2c_speed[False-backward] 6.9494ms 6.4113ms 155.9752 Ops/s 161.2159 Ops/s $\color{#d91a1a}-3.25\%$
test_a2c_speed[True-None] 1.4985ms 1.3450ms 743.4782 Ops/s 739.8950 Ops/s $\color{#35bf28}+0.48\%$
test_a2c_speed[True-backward] 3.1685ms 3.0397ms 328.9837 Ops/s 321.6118 Ops/s $\color{#35bf28}+2.29\%$
test_a2c_speed[reduce-overhead-None] 15.9956ms 9.0796ms 110.1367 Ops/s 110.8098 Ops/s $\color{#d91a1a}-0.61\%$
test_a2c_speed[reduce-overhead-backward] 2.2894ms 1.7579ms 568.8511 Ops/s 610.9015 Ops/s $\textbf{\color{#d91a1a}-6.88\%}$
test_ppo_speed[False-None] 4.3524ms 3.8425ms 260.2498 Ops/s 266.4066 Ops/s $\color{#d91a1a}-2.31\%$
test_ppo_speed[False-backward] 7.6976ms 7.1991ms 138.9069 Ops/s 141.4809 Ops/s $\color{#d91a1a}-1.82\%$
test_ppo_speed[True-None] 1.6374ms 1.4108ms 708.8022 Ops/s 700.2005 Ops/s $\color{#35bf28}+1.23\%$
test_ppo_speed[True-backward] 3.3382ms 3.0542ms 327.4229 Ops/s 320.4727 Ops/s $\color{#35bf28}+2.17\%$
test_ppo_speed[reduce-overhead-None] 1.1278ms 0.9721ms 1.0287 KOps/s 1.0227 KOps/s $\color{#35bf28}+0.58\%$
test_ppo_speed[reduce-overhead-backward] 1.5460ms 1.4076ms 710.4492 Ops/s 687.1470 Ops/s $\color{#35bf28}+3.39\%$
test_reinforce_speed[False-None] 2.8978ms 2.3285ms 429.4519 Ops/s 433.5765 Ops/s $\color{#d91a1a}-0.95\%$
test_reinforce_speed[False-backward] 3.9223ms 3.3552ms 298.0422 Ops/s 301.7378 Ops/s $\color{#d91a1a}-1.22\%$
test_reinforce_speed[True-None] 1.4346ms 1.2902ms 775.0990 Ops/s 764.1021 Ops/s $\color{#35bf28}+1.44\%$
test_reinforce_speed[True-backward] 3.2271ms 2.9436ms 339.7156 Ops/s 340.7025 Ops/s $\color{#d91a1a}-0.29\%$
test_reinforce_speed[reduce-overhead-None] 18.1518ms 10.0869ms 99.1389 Ops/s 101.5126 Ops/s $\color{#d91a1a}-2.34\%$
test_reinforce_speed[reduce-overhead-backward] 1.5404ms 1.4672ms 681.5801 Ops/s 653.5613 Ops/s $\color{#35bf28}+4.29\%$
test_iql_speed[False-None] 9.6933ms 9.3122ms 107.3856 Ops/s 105.9714 Ops/s $\color{#35bf28}+1.33\%$
test_iql_speed[False-backward] 13.5124ms 12.9591ms 77.1659 Ops/s 76.4002 Ops/s $\color{#35bf28}+1.00\%$
test_iql_speed[True-None] 2.5497ms 2.2427ms 445.8969 Ops/s 433.5925 Ops/s $\color{#35bf28}+2.84\%$
test_iql_speed[True-backward] 5.3361ms 4.7973ms 208.4488 Ops/s 199.4387 Ops/s $\color{#35bf28}+4.52\%$
test_iql_speed[reduce-overhead-None] 19.0345ms 11.2310ms 89.0396 Ops/s 90.1526 Ops/s $\color{#d91a1a}-1.23\%$
test_iql_speed[reduce-overhead-backward] 2.0873ms 1.9012ms 525.9825 Ops/s 500.1988 Ops/s $\textbf{\color{#35bf28}+5.15\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.8248ms 6.2987ms 158.7622 Ops/s 154.1517 Ops/s $\color{#35bf28}+2.99\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6579ms 0.3919ms 2.5519 KOps/s 3.3180 KOps/s $\textbf{\color{#d91a1a}-23.09\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7050ms 0.3743ms 2.6719 KOps/s 3.8172 KOps/s $\textbf{\color{#d91a1a}-30.00\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.5379ms 6.0655ms 164.8656 Ops/s 160.7915 Ops/s $\color{#35bf28}+2.53\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2504ms 0.3358ms 2.9779 KOps/s 3.4203 KOps/s $\textbf{\color{#d91a1a}-12.93\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6444ms 0.3206ms 3.1188 KOps/s 3.5794 KOps/s $\textbf{\color{#d91a1a}-12.87\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6804ms 1.4190ms 704.7219 Ops/s 716.7226 Ops/s $\color{#d91a1a}-1.67\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4635ms 1.2193ms 820.1699 Ops/s 752.5415 Ops/s $\textbf{\color{#35bf28}+8.99\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5948ms 6.2634ms 159.6580 Ops/s 156.5250 Ops/s $\color{#35bf28}+2.00\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9583ms 0.4243ms 2.3567 KOps/s 2.2945 KOps/s $\color{#35bf28}+2.71\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6903ms 0.4380ms 2.2833 KOps/s 2.2040 KOps/s $\color{#35bf28}+3.60\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.7357ms 6.1267ms 163.2200 Ops/s 160.9056 Ops/s $\color{#35bf28}+1.44\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8303ms 0.2935ms 3.4074 KOps/s 3.0063 KOps/s $\textbf{\color{#35bf28}+13.34\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7383ms 0.3389ms 2.9505 KOps/s 3.2984 KOps/s $\textbf{\color{#d91a1a}-10.55\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 10.2031ms 6.1667ms 162.1617 Ops/s 162.4194 Ops/s $\color{#d91a1a}-0.16\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8996ms 0.3329ms 3.0040 KOps/s 3.2543 KOps/s $\textbf{\color{#d91a1a}-7.69\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5456ms 0.3084ms 3.2428 KOps/s 3.0231 KOps/s $\textbf{\color{#35bf28}+7.27\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6060ms 6.2383ms 160.3003 Ops/s 156.3047 Ops/s $\color{#35bf28}+2.56\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1536ms 0.4451ms 2.2468 KOps/s 2.0889 KOps/s $\textbf{\color{#35bf28}+7.56\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7229ms 0.4353ms 2.2974 KOps/s 2.0534 KOps/s $\textbf{\color{#35bf28}+11.88\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0781ms 5.5432ms 180.4028 Ops/s 177.0301 Ops/s $\color{#35bf28}+1.91\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.1200ms 2.0677ms 483.6261 Ops/s 428.3700 Ops/s $\textbf{\color{#35bf28}+12.90\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.1228ms 1.2143ms 823.5502 Ops/s 790.6449 Ops/s $\color{#35bf28}+4.16\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.1252ms 5.6330ms 177.5266 Ops/s 177.8974 Ops/s $\color{#d91a1a}-0.21\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.5639ms 2.0671ms 483.7693 Ops/s 414.0695 Ops/s $\textbf{\color{#35bf28}+16.83\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1832ms 1.2198ms 819.8198 Ops/s 825.1362 Ops/s $\color{#d91a1a}-0.64\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4954s 15.5959ms 64.1194 Ops/s 31.4044 Ops/s $\textbf{\color{#35bf28}+104.17\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.9772ms 2.3026ms 434.2854 Ops/s 451.1766 Ops/s $\color{#d91a1a}-3.74\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 9.6052ms 1.5099ms 662.3094 Ops/s 733.8907 Ops/s $\textbf{\color{#d91a1a}-9.75\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.2379ms 12.9346ms 77.3119 Ops/s 73.2053 Ops/s $\textbf{\color{#35bf28}+5.61\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.2798ms 16.8100ms 59.4885 Ops/s 58.9290 Ops/s $\color{#35bf28}+0.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.0744ms 17.6923ms 56.5217 Ops/s 55.4873 Ops/s $\color{#35bf28}+1.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7884ms 17.0046ms 58.8076 Ops/s 58.2023 Ops/s $\color{#35bf28}+1.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 17.9462ms 17.6363ms 56.7014 Ops/s 54.4674 Ops/s $\color{#35bf28}+4.10\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.5706ms 18.3290ms 54.5583 Ops/s 53.2647 Ops/s $\color{#35bf28}+2.43\%$

[ghstack-poisoned]
@vmoens vmoens added bc breaking backward compatibility breaking change Deprecation labels Feb 4, 2025
@vmoens vmoens merged commit 650cf9a into gh/vmoens/90/base Feb 4, 2025
27 of 50 checks passed
vmoens added a commit that referenced this pull request Feb 4, 2025
ghstack-source-id: 4a70178f54f9e25d602c86a0b61248d66f3e39bd
Pull Request resolved: #2747
@vmoens vmoens deleted the gh/vmoens/90/head branch February 4, 2025 08:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bc breaking backward compatibility breaking change CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Deprecation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants