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 AdditiveGaussianWrapper #2748

Merged
merged 3 commits into from
Feb 4, 2025

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/2748

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.

[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: $\large\color{#35bf28}18$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5771s 0.4740s 2.1098 Ops/s 2.1993 Ops/s $\color{#d91a1a}-4.07\%$
test_transformed 1.0731s 0.9752s 1.0254 Ops/s 1.0910 Ops/s $\textbf{\color{#d91a1a}-6.01\%}$
test_serial 1.5350s 1.4313s 0.6987 Ops/s 0.7198 Ops/s $\color{#d91a1a}-2.94\%$
test_parallel 1.3473s 1.2438s 0.8040 Ops/s 0.8222 Ops/s $\color{#d91a1a}-2.21\%$
test_step_mdp_speed[True-True-True-True-True] 0.1529ms 30.1347μs 33.1843 KOps/s 33.3137 KOps/s $\color{#d91a1a}-0.39\%$
test_step_mdp_speed[True-True-True-True-False] 51.2250μs 17.6525μs 56.6493 KOps/s 56.5870 KOps/s $\color{#35bf28}+0.11\%$
test_step_mdp_speed[True-True-True-False-True] 45.7050μs 17.1295μs 58.3789 KOps/s 58.7527 KOps/s $\color{#d91a1a}-0.64\%$
test_step_mdp_speed[True-True-True-False-False] 63.0170μs 10.0987μs 99.0229 KOps/s 100.8460 KOps/s $\color{#d91a1a}-1.81\%$
test_step_mdp_speed[True-True-False-True-True] 75.6910μs 32.0336μs 31.2173 KOps/s 30.8604 KOps/s $\color{#35bf28}+1.16\%$
test_step_mdp_speed[True-True-False-True-False] 67.5860μs 19.4437μs 51.4305 KOps/s 51.6006 KOps/s $\color{#d91a1a}-0.33\%$
test_step_mdp_speed[True-True-False-False-True] 50.7840μs 18.9097μs 52.8828 KOps/s 53.1941 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[True-True-False-False-False] 42.5190μs 11.7543μs 85.0751 KOps/s 84.4229 KOps/s $\color{#35bf28}+0.77\%$
test_step_mdp_speed[True-False-True-True-True] 93.9530μs 33.8930μs 29.5047 KOps/s 29.5173 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[True-False-True-True-False] 68.0770μs 21.7438μs 45.9901 KOps/s 47.1273 KOps/s $\color{#d91a1a}-2.41\%$
test_step_mdp_speed[True-False-True-False-True] 60.8440μs 18.7918μs 53.2147 KOps/s 51.2931 KOps/s $\color{#35bf28}+3.75\%$
test_step_mdp_speed[True-False-True-False-False] 0.5532ms 11.9342μs 83.7927 KOps/s 84.9076 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[True-False-False-True-True] 0.1089ms 35.6252μs 28.0700 KOps/s 28.1239 KOps/s $\color{#d91a1a}-0.19\%$
test_step_mdp_speed[True-False-False-True-False] 66.4640μs 23.3111μs 42.8980 KOps/s 43.1842 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[True-False-False-False-True] 81.5520μs 20.5282μs 48.7135 KOps/s 48.6393 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[True-False-False-False-False] 58.9590μs 13.5207μs 73.9605 KOps/s 73.5829 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[False-True-True-True-True] 80.0190μs 33.8158μs 29.5719 KOps/s 29.4907 KOps/s $\color{#35bf28}+0.28\%$
test_step_mdp_speed[False-True-True-True-False] 98.6040μs 21.5303μs 46.4463 KOps/s 47.1398 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-True-True-False-True] 67.7860μs 21.7454μs 45.9868 KOps/s 46.3923 KOps/s $\color{#d91a1a}-0.87\%$
test_step_mdp_speed[False-True-True-False-False] 63.0880μs 13.1811μs 75.8661 KOps/s 75.9567 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-True-False-True-True] 92.3220μs 35.5064μs 28.1639 KOps/s 27.9930 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[False-True-False-True-False] 65.4920μs 23.1424μs 43.2107 KOps/s 43.5981 KOps/s $\color{#d91a1a}-0.89\%$
test_step_mdp_speed[False-True-False-False-True] 2.7920ms 23.2523μs 43.0065 KOps/s 43.1619 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[False-True-False-False-False] 58.1880μs 15.0603μs 66.3998 KOps/s 67.1649 KOps/s $\color{#d91a1a}-1.14\%$
test_step_mdp_speed[False-False-True-True-True] 0.1332ms 36.9945μs 27.0310 KOps/s 26.4963 KOps/s $\color{#35bf28}+2.02\%$
test_step_mdp_speed[False-False-True-True-False] 79.3980μs 25.0599μs 39.9044 KOps/s 39.8311 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-False-True-False-True] 79.3180μs 23.1478μs 43.2007 KOps/s 43.5658 KOps/s $\color{#d91a1a}-0.84\%$
test_step_mdp_speed[False-False-True-False-False] 54.0710μs 14.9520μs 66.8805 KOps/s 67.8680 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[False-False-False-True-True] 74.0280μs 39.1659μs 25.5324 KOps/s 25.9007 KOps/s $\color{#d91a1a}-1.42\%$
test_step_mdp_speed[False-False-False-True-False] 0.7125ms 26.7778μs 37.3443 KOps/s 37.1494 KOps/s $\color{#35bf28}+0.52\%$
test_step_mdp_speed[False-False-False-False-True] 86.8820μs 24.7040μs 40.4793 KOps/s 40.9434 KOps/s $\color{#d91a1a}-1.13\%$
test_step_mdp_speed[False-False-False-False-False] 44.4320μs 16.7979μs 59.5314 KOps/s 61.0110 KOps/s $\color{#d91a1a}-2.43\%$
test_values[generalized_advantage_estimate-True-True] 12.2782ms 9.9244ms 100.7621 Ops/s 104.5716 Ops/s $\color{#d91a1a}-3.64\%$
test_values[vec_generalized_advantage_estimate-True-True] 25.3850ms 24.4072ms 40.9715 Ops/s 41.2531 Ops/s $\color{#d91a1a}-0.68\%$
test_values[td0_return_estimate-False-False] 0.2424ms 0.1869ms 5.3503 KOps/s 5.0356 KOps/s $\textbf{\color{#35bf28}+6.25\%}$
test_values[td1_return_estimate-False-False] 35.0600ms 25.3308ms 39.4776 Ops/s 41.8789 Ops/s $\textbf{\color{#d91a1a}-5.73\%}$
test_values[vec_td1_return_estimate-False-False] 27.2516ms 24.5661ms 40.7066 Ops/s 40.6077 Ops/s $\color{#35bf28}+0.24\%$
test_values[td_lambda_return_estimate-True-False] 38.9438ms 35.5615ms 28.1203 Ops/s 28.6580 Ops/s $\color{#d91a1a}-1.88\%$
test_values[vec_td_lambda_return_estimate-True-False] 26.6995ms 24.6331ms 40.5958 Ops/s 40.5720 Ops/s $\color{#35bf28}+0.06\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.0328ms 8.5419ms 117.0696 Ops/s 118.9764 Ops/s $\color{#d91a1a}-1.60\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4125ms 1.9476ms 513.4506 Ops/s 488.1931 Ops/s $\textbf{\color{#35bf28}+5.17\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5756ms 0.3755ms 2.6630 KOps/s 2.6557 KOps/s $\color{#35bf28}+0.28\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 44.1375ms 41.4033ms 24.1527 Ops/s 21.6490 Ops/s $\textbf{\color{#35bf28}+11.56\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.5023ms 3.4788ms 287.4520 Ops/s 287.6091 Ops/s $\color{#d91a1a}-0.05\%$
test_dqn_speed[False-None] 2.0523ms 1.4351ms 696.8129 Ops/s 695.7055 Ops/s $\color{#35bf28}+0.16\%$
test_dqn_speed[False-backward] 2.0672ms 1.9366ms 516.3706 Ops/s 514.8344 Ops/s $\color{#35bf28}+0.30\%$
test_dqn_speed[True-None] 0.6149ms 0.4854ms 2.0600 KOps/s 2.0462 KOps/s $\color{#35bf28}+0.67\%$
test_dqn_speed[True-backward] 0.9605ms 0.9083ms 1.1009 KOps/s 1.0693 KOps/s $\color{#35bf28}+2.96\%$
test_dqn_speed[reduce-overhead-None] 0.8466ms 0.4964ms 2.0145 KOps/s 2.0146 KOps/s $-0.00\%$
test_dqn_speed[reduce-overhead-backward] 1.0027ms 0.9223ms 1.0842 KOps/s 1.0799 KOps/s $\color{#35bf28}+0.40\%$
test_ddpg_speed[False-None] 3.4136ms 2.9810ms 335.4606 Ops/s 336.9139 Ops/s $\color{#d91a1a}-0.43\%$
test_ddpg_speed[False-backward] 4.4420ms 4.1755ms 239.4941 Ops/s 240.2859 Ops/s $\color{#d91a1a}-0.33\%$
test_ddpg_speed[True-None] 1.4733ms 1.2413ms 805.6123 Ops/s 800.6900 Ops/s $\color{#35bf28}+0.61\%$
test_ddpg_speed[True-backward] 2.2338ms 2.1459ms 466.0056 Ops/s 417.2686 Ops/s $\textbf{\color{#35bf28}+11.68\%}$
test_ddpg_speed[reduce-overhead-None] 1.4243ms 1.2374ms 808.1716 Ops/s 799.1363 Ops/s $\color{#35bf28}+1.13\%$
test_ddpg_speed[reduce-overhead-backward] 2.1914ms 2.1271ms 470.1212 Ops/s 468.7677 Ops/s $\color{#35bf28}+0.29\%$
test_sac_speed[False-None] 9.5810ms 8.1972ms 121.9923 Ops/s 124.4988 Ops/s $\color{#d91a1a}-2.01\%$
test_sac_speed[False-backward] 12.4525ms 10.9173ms 91.5977 Ops/s 92.3389 Ops/s $\color{#d91a1a}-0.80\%$
test_sac_speed[True-None] 2.6570ms 2.0852ms 479.5642 Ops/s 472.2160 Ops/s $\color{#35bf28}+1.56\%$
test_sac_speed[True-backward] 4.2205ms 3.8534ms 259.5125 Ops/s 263.4163 Ops/s $\color{#d91a1a}-1.48\%$
test_sac_speed[reduce-overhead-None] 2.3302ms 2.0764ms 481.5914 Ops/s 475.7129 Ops/s $\color{#35bf28}+1.24\%$
test_sac_speed[reduce-overhead-backward] 4.0521ms 3.7625ms 265.7811 Ops/s 262.3826 Ops/s $\color{#35bf28}+1.30\%$
test_redq_speed[False-None] 14.0847ms 13.0937ms 76.3726 Ops/s 77.6338 Ops/s $\color{#d91a1a}-1.62\%$
test_redq_speed[False-backward] 23.4098ms 22.5430ms 44.3596 Ops/s 45.0642 Ops/s $\color{#d91a1a}-1.56\%$
test_redq_speed[True-None] 5.6478ms 5.1799ms 193.0556 Ops/s 194.5203 Ops/s $\color{#d91a1a}-0.75\%$
test_redq_speed[True-backward] 14.8417ms 12.8045ms 78.0975 Ops/s 81.8559 Ops/s $\color{#d91a1a}-4.59\%$
test_redq_speed[reduce-overhead-None] 6.0509ms 5.6055ms 178.3966 Ops/s 186.0044 Ops/s $\color{#d91a1a}-4.09\%$
test_redq_speed[reduce-overhead-backward] 14.0820ms 12.5399ms 79.7452 Ops/s 74.4339 Ops/s $\textbf{\color{#35bf28}+7.14\%}$
test_redq_deprec_speed[False-None] 14.4945ms 13.0052ms 76.8922 Ops/s 74.9606 Ops/s $\color{#35bf28}+2.58\%$
test_redq_deprec_speed[False-backward] 20.5308ms 18.5608ms 53.8771 Ops/s 52.2153 Ops/s $\color{#35bf28}+3.18\%$
test_redq_deprec_speed[True-None] 4.9938ms 3.8176ms 261.9460 Ops/s 225.8058 Ops/s $\textbf{\color{#35bf28}+16.01\%}$
test_redq_deprec_speed[True-backward] 9.4810ms 9.0699ms 110.2554 Ops/s 105.4095 Ops/s $\color{#35bf28}+4.60\%$
test_redq_deprec_speed[reduce-overhead-None] 4.2805ms 3.7893ms 263.9027 Ops/s 257.1975 Ops/s $\color{#35bf28}+2.61\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.9153ms 8.1745ms 122.3309 Ops/s 118.5402 Ops/s $\color{#35bf28}+3.20\%$
test_td3_speed[False-None] 8.9285ms 8.3170ms 120.2353 Ops/s 120.5354 Ops/s $\color{#d91a1a}-0.25\%$
test_td3_speed[False-backward] 12.9179ms 11.1683ms 89.5388 Ops/s 93.3935 Ops/s $\color{#d91a1a}-4.13\%$
test_td3_speed[True-None] 2.0016ms 1.8350ms 544.9668 Ops/s 552.3769 Ops/s $\color{#d91a1a}-1.34\%$
test_td3_speed[True-backward] 4.1688ms 3.7565ms 266.2086 Ops/s 273.8687 Ops/s $\color{#d91a1a}-2.80\%$
test_td3_speed[reduce-overhead-None] 1.9899ms 1.7815ms 561.3317 Ops/s 551.2086 Ops/s $\color{#35bf28}+1.84\%$
test_td3_speed[reduce-overhead-backward] 3.5107ms 3.4133ms 292.9758 Ops/s 292.7857 Ops/s $\color{#35bf28}+0.06\%$
test_cql_speed[False-None] 39.7086ms 37.1623ms 26.9090 Ops/s 26.6351 Ops/s $\color{#35bf28}+1.03\%$
test_cql_speed[False-backward] 48.7524ms 47.1173ms 21.2236 Ops/s 20.6048 Ops/s $\color{#35bf28}+3.00\%$
test_cql_speed[True-None] 17.5979ms 16.3164ms 61.2881 Ops/s 61.3043 Ops/s $\color{#d91a1a}-0.03\%$
test_cql_speed[True-backward] 24.4788ms 23.1569ms 43.1836 Ops/s 43.7253 Ops/s $\color{#d91a1a}-1.24\%$
test_cql_speed[reduce-overhead-None] 17.8400ms 16.6101ms 60.2045 Ops/s 62.6679 Ops/s $\color{#d91a1a}-3.93\%$
test_cql_speed[reduce-overhead-backward] 29.9623ms 23.6279ms 42.3229 Ops/s 43.6126 Ops/s $\color{#d91a1a}-2.96\%$
test_a2c_speed[False-None] 8.8445ms 7.3295ms 136.4342 Ops/s 138.8538 Ops/s $\color{#d91a1a}-1.74\%$
test_a2c_speed[False-backward] 15.0133ms 14.5166ms 68.8867 Ops/s 69.7925 Ops/s $\color{#d91a1a}-1.30\%$
test_a2c_speed[True-None] 4.4352ms 3.6936ms 270.7375 Ops/s 266.4611 Ops/s $\color{#35bf28}+1.60\%$
test_a2c_speed[True-backward] 11.3885ms 10.2991ms 97.0961 Ops/s 91.9151 Ops/s $\textbf{\color{#35bf28}+5.64\%}$
test_a2c_speed[reduce-overhead-None] 4.0937ms 3.6917ms 270.8758 Ops/s 258.9360 Ops/s $\color{#35bf28}+4.61\%$
test_a2c_speed[reduce-overhead-backward] 10.5188ms 10.1143ms 98.8697 Ops/s 96.6322 Ops/s $\color{#35bf28}+2.32\%$
test_ppo_speed[False-None] 8.3159ms 7.5909ms 131.7359 Ops/s 132.1667 Ops/s $\color{#d91a1a}-0.33\%$
test_ppo_speed[False-backward] 16.9526ms 15.7741ms 63.3949 Ops/s 67.1926 Ops/s $\textbf{\color{#d91a1a}-5.65\%}$
test_ppo_speed[True-None] 4.4884ms 4.0925ms 244.3506 Ops/s 209.3380 Ops/s $\textbf{\color{#35bf28}+16.73\%}$
test_ppo_speed[True-backward] 10.6019ms 10.1264ms 98.7521 Ops/s 97.3881 Ops/s $\color{#35bf28}+1.40\%$
test_ppo_speed[reduce-overhead-None] 4.4240ms 4.1067ms 243.5060 Ops/s 241.5622 Ops/s $\color{#35bf28}+0.80\%$
test_ppo_speed[reduce-overhead-backward] 10.4570ms 10.0822ms 99.1846 Ops/s 96.3482 Ops/s $\color{#35bf28}+2.94\%$
test_reinforce_speed[False-None] 7.9534ms 6.6178ms 151.1070 Ops/s 149.9671 Ops/s $\color{#35bf28}+0.76\%$
test_reinforce_speed[False-backward] 10.3646ms 10.0546ms 99.4565 Ops/s 97.6389 Ops/s $\color{#35bf28}+1.86\%$
test_reinforce_speed[True-None] 3.3944ms 3.0923ms 323.3834 Ops/s 326.3754 Ops/s $\color{#d91a1a}-0.92\%$
test_reinforce_speed[True-backward] 10.1616ms 9.7513ms 102.5507 Ops/s 107.5045 Ops/s $\color{#d91a1a}-4.61\%$
test_reinforce_speed[reduce-overhead-None] 3.7865ms 3.1217ms 320.3423 Ops/s 314.0004 Ops/s $\color{#35bf28}+2.02\%$
test_reinforce_speed[reduce-overhead-backward] 11.1216ms 9.9126ms 100.8813 Ops/s 107.2039 Ops/s $\textbf{\color{#d91a1a}-5.90\%}$
test_iql_speed[False-None] 35.5093ms 34.2378ms 29.2075 Ops/s 29.2897 Ops/s $\color{#d91a1a}-0.28\%$
test_iql_speed[False-backward] 49.4393ms 47.7911ms 20.9244 Ops/s 21.3486 Ops/s $\color{#d91a1a}-1.99\%$
test_iql_speed[True-None] 13.2303ms 11.9031ms 84.0118 Ops/s 84.7585 Ops/s $\color{#d91a1a}-0.88\%$
test_iql_speed[True-backward] 23.4458ms 22.6962ms 44.0603 Ops/s 42.5874 Ops/s $\color{#35bf28}+3.46\%$
test_iql_speed[reduce-overhead-None] 12.8856ms 11.9453ms 83.7149 Ops/s 84.5746 Ops/s $\color{#d91a1a}-1.02\%$
test_iql_speed[reduce-overhead-backward] 23.9557ms 22.7230ms 44.0083 Ops/s 42.7532 Ops/s $\color{#35bf28}+2.94\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.3702ms 4.9180ms 203.3354 Ops/s 189.6207 Ops/s $\textbf{\color{#35bf28}+7.23\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8930ms 0.5254ms 1.9031 KOps/s 1.8622 KOps/s $\color{#35bf28}+2.20\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9553ms 0.5087ms 1.9658 KOps/s 1.9773 KOps/s $\color{#d91a1a}-0.58\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.5822ms 4.7560ms 210.2623 Ops/s 201.0711 Ops/s $\color{#35bf28}+4.57\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0039ms 0.5115ms 1.9550 KOps/s 1.8935 KOps/s $\color{#35bf28}+3.25\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8341ms 0.4923ms 2.0314 KOps/s 1.9898 KOps/s $\color{#35bf28}+2.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.9458ms 1.6650ms 600.6084 Ops/s 589.2502 Ops/s $\color{#35bf28}+1.93\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.2868ms 1.5797ms 633.0457 Ops/s 621.4996 Ops/s $\color{#35bf28}+1.86\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.6799ms 4.9929ms 200.2830 Ops/s 187.0058 Ops/s $\textbf{\color{#35bf28}+7.10\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5107ms 0.6753ms 1.4808 KOps/s 1.4559 KOps/s $\color{#35bf28}+1.71\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9462ms 0.6404ms 1.5616 KOps/s 1.5383 KOps/s $\color{#35bf28}+1.52\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9648ms 4.8942ms 204.3216 Ops/s 191.2873 Ops/s $\textbf{\color{#35bf28}+6.81\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9282ms 0.5309ms 1.8836 KOps/s 1.8289 KOps/s $\color{#35bf28}+2.99\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8187ms 0.5104ms 1.9593 KOps/s 1.9485 KOps/s $\color{#35bf28}+0.55\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.1502ms 4.8448ms 206.4052 Ops/s 199.5859 Ops/s $\color{#35bf28}+3.42\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9418ms 0.5287ms 1.8913 KOps/s 1.8670 KOps/s $\color{#35bf28}+1.30\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6852ms 0.5008ms 1.9967 KOps/s 1.9422 KOps/s $\color{#35bf28}+2.80\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.6492ms 4.9888ms 200.4509 Ops/s 188.1972 Ops/s $\textbf{\color{#35bf28}+6.51\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.4953s 1.3573ms 736.7408 Ops/s 1.4597 KOps/s $\textbf{\color{#d91a1a}-49.53\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9369ms 0.6294ms 1.5887 KOps/s 1.5152 KOps/s $\color{#35bf28}+4.85\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.6641ms 4.1607ms 240.3468 Ops/s 238.6876 Ops/s $\color{#35bf28}+0.70\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.3535ms 2.4186ms 413.4704 Ops/s 421.6794 Ops/s $\color{#d91a1a}-1.95\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.7780ms 1.2437ms 804.0440 Ops/s 733.6280 Ops/s $\textbf{\color{#35bf28}+9.60\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 5.4287ms 4.1875ms 238.8035 Ops/s 31.1445 Ops/s $\textbf{\color{#35bf28}+666.76\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.9323ms 2.3638ms 423.0539 Ops/s 401.3501 Ops/s $\textbf{\color{#35bf28}+5.41\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1099ms 1.4694ms 680.5322 Ops/s 765.7781 Ops/s $\textbf{\color{#d91a1a}-11.13\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4321s 13.0146ms 76.8370 Ops/s 211.5616 Ops/s $\textbf{\color{#d91a1a}-63.68\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.8595ms 2.5866ms 386.6131 Ops/s 389.9190 Ops/s $\color{#d91a1a}-0.85\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.2396ms 1.5586ms 641.5879 Ops/s 595.3601 Ops/s $\textbf{\color{#35bf28}+7.76\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 11.9447ms 11.6313ms 85.9746 Ops/s 80.8186 Ops/s $\textbf{\color{#35bf28}+6.38\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 15.8873ms 14.5178ms 68.8811 Ops/s 68.2375 Ops/s $\color{#35bf28}+0.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 21.6895ms 20.4124ms 48.9898 Ops/s 47.3407 Ops/s $\color{#35bf28}+3.48\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.9053ms 14.7067ms 67.9963 Ops/s 67.4135 Ops/s $\color{#35bf28}+0.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.7351ms 20.3149ms 49.2250 Ops/s 46.8416 Ops/s $\textbf{\color{#35bf28}+5.09\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.6254ms 15.7900ms 63.3313 Ops/s 61.7171 Ops/s $\color{#35bf28}+2.62\%$

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: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8371s 0.7493s 1.3346 Ops/s 1.3766 Ops/s $\color{#d91a1a}-3.05\%$
test_transformed 1.4493s 1.3579s 0.7364 Ops/s 0.7679 Ops/s $\color{#d91a1a}-4.10\%$
test_serial 2.2486s 2.2001s 0.4545 Ops/s 0.4612 Ops/s $\color{#d91a1a}-1.44\%$
test_parallel 1.8802s 1.8284s 0.5469 Ops/s 0.5375 Ops/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[True-True-True-True-True] 0.2091ms 39.4073μs 25.3760 KOps/s 25.7320 KOps/s $\color{#d91a1a}-1.38\%$
test_step_mdp_speed[True-True-True-True-False] 0.4034ms 23.6378μs 42.3051 KOps/s 42.0090 KOps/s $\color{#35bf28}+0.70\%$
test_step_mdp_speed[True-True-True-False-True] 54.3610μs 22.0777μs 45.2946 KOps/s 44.5983 KOps/s $\color{#35bf28}+1.56\%$
test_step_mdp_speed[True-True-True-False-False] 0.4062ms 13.0026μs 76.9079 KOps/s 76.6190 KOps/s $\color{#35bf28}+0.38\%$
test_step_mdp_speed[True-True-False-True-True] 0.4435ms 42.8793μs 23.3213 KOps/s 23.7018 KOps/s $\color{#d91a1a}-1.61\%$
test_step_mdp_speed[True-True-False-True-False] 62.2610μs 25.8635μs 38.6646 KOps/s 39.0198 KOps/s $\color{#d91a1a}-0.91\%$
test_step_mdp_speed[True-True-False-False-True] 0.4174ms 24.4507μs 40.8986 KOps/s 40.3887 KOps/s $\color{#35bf28}+1.26\%$
test_step_mdp_speed[True-True-False-False-False] 0.4110ms 15.4073μs 64.9045 KOps/s 64.7301 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-False-True-True-True] 0.4385ms 45.5163μs 21.9702 KOps/s 22.1810 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[True-False-True-True-False] 59.8310μs 28.3083μs 35.3254 KOps/s 35.6652 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[True-False-True-False-True] 0.4295ms 25.0107μs 39.9828 KOps/s 41.7213 KOps/s $\color{#d91a1a}-4.17\%$
test_step_mdp_speed[True-False-True-False-False] 0.4200ms 15.5524μs 64.2988 KOps/s 66.3868 KOps/s $\color{#d91a1a}-3.15\%$
test_step_mdp_speed[True-False-False-True-True] 0.4494ms 47.2372μs 21.1698 KOps/s 21.3721 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[True-False-False-True-False] 69.8210μs 30.2246μs 33.0856 KOps/s 33.3061 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[True-False-False-False-True] 0.4159ms 26.6023μs 37.5907 KOps/s 37.5023 KOps/s $\color{#35bf28}+0.24\%$
test_step_mdp_speed[True-False-False-False-False] 0.4186ms 17.5338μs 57.0327 KOps/s 57.6662 KOps/s $\color{#d91a1a}-1.10\%$
test_step_mdp_speed[False-True-True-True-True] 0.4571ms 44.5468μs 22.4483 KOps/s 22.4111 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[False-True-True-True-False] 61.4210μs 27.9762μs 35.7447 KOps/s 35.7005 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[False-True-True-False-True] 2.7810ms 28.8843μs 34.6209 KOps/s 35.2695 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[False-True-True-False-False] 0.4336ms 17.2221μs 58.0650 KOps/s 59.3374 KOps/s $\color{#d91a1a}-2.14\%$
test_step_mdp_speed[False-True-False-True-True] 0.4543ms 46.8782μs 21.3319 KOps/s 21.5562 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[False-True-False-True-False] 64.9220μs 30.1545μs 33.1625 KOps/s 33.0975 KOps/s $\color{#35bf28}+0.20\%$
test_step_mdp_speed[False-True-False-False-True] 0.4246ms 30.8692μs 32.3948 KOps/s 31.4636 KOps/s $\color{#35bf28}+2.96\%$
test_step_mdp_speed[False-True-False-False-False] 51.0010μs 19.2852μs 51.8532 KOps/s 51.7613 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-False-True-True-True] 0.4410ms 49.0768μs 20.3762 KOps/s 19.9650 KOps/s $\color{#35bf28}+2.06\%$
test_step_mdp_speed[False-False-True-True-False] 0.4224ms 32.7334μs 30.5499 KOps/s 30.5590 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[False-False-True-False-True] 85.7910μs 30.6081μs 32.6711 KOps/s 32.7380 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[False-False-True-False-False] 0.4038ms 19.2254μs 52.0145 KOps/s 52.8313 KOps/s $\color{#d91a1a}-1.55\%$
test_step_mdp_speed[False-False-False-True-True] 0.4431ms 51.3099μs 19.4894 KOps/s 19.4548 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-False-False-True-False] 0.4285ms 34.7738μs 28.7573 KOps/s 29.0685 KOps/s $\color{#d91a1a}-1.07\%$
test_step_mdp_speed[False-False-False-False-True] 65.6310μs 31.8295μs 31.4174 KOps/s 30.5223 KOps/s $\color{#35bf28}+2.93\%$
test_step_mdp_speed[False-False-False-False-False] 0.4185ms 21.3469μs 46.8453 KOps/s 46.8404 KOps/s $\color{#35bf28}+0.01\%$
test_values[generalized_advantage_estimate-True-True] 25.7487ms 25.2165ms 39.6566 Ops/s 40.4096 Ops/s $\color{#d91a1a}-1.86\%$
test_values[vec_generalized_advantage_estimate-True-True] 96.8237ms 2.8361ms 352.5912 Ops/s 317.3436 Ops/s $\textbf{\color{#35bf28}+11.11\%}$
test_values[td0_return_estimate-False-False] 0.1084ms 80.2832μs 12.4559 KOps/s 12.7003 KOps/s $\color{#d91a1a}-1.92\%$
test_values[td1_return_estimate-False-False] 56.5161ms 55.9399ms 17.8763 Ops/s 18.1896 Ops/s $\color{#d91a1a}-1.72\%$
test_values[vec_td1_return_estimate-False-False] 1.3998ms 1.0892ms 918.0934 Ops/s 924.9776 Ops/s $\color{#d91a1a}-0.74\%$
test_values[td_lambda_return_estimate-True-False] 89.5812ms 88.5377ms 11.2946 Ops/s 11.4998 Ops/s $\color{#d91a1a}-1.78\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4697ms 1.0930ms 914.9447 Ops/s 930.5105 Ops/s $\color{#d91a1a}-1.67\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 27.2949ms 26.3148ms 38.0014 Ops/s 40.3349 Ops/s $\textbf{\color{#d91a1a}-5.79\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.1038ms 0.7595ms 1.3167 KOps/s 1.3236 KOps/s $\color{#d91a1a}-0.53\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8159ms 0.6734ms 1.4851 KOps/s 1.4992 KOps/s $\color{#d91a1a}-0.94\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6378ms 1.4884ms 671.8609 Ops/s 674.0435 Ops/s $\color{#d91a1a}-0.32\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8721ms 0.7248ms 1.3797 KOps/s 1.4683 KOps/s $\textbf{\color{#d91a1a}-6.04\%}$
test_dqn_speed[False-None] 1.7222ms 1.5277ms 654.5669 Ops/s 657.8364 Ops/s $\color{#d91a1a}-0.50\%$
test_dqn_speed[False-backward] 2.3714ms 2.1633ms 462.2573 Ops/s 474.1442 Ops/s $\color{#d91a1a}-2.51\%$
test_dqn_speed[True-None] 0.9598ms 0.5557ms 1.7994 KOps/s 1.7772 KOps/s $\color{#35bf28}+1.25\%$
test_dqn_speed[True-backward] 1.2967ms 1.2264ms 815.4126 Ops/s 810.3634 Ops/s $\color{#35bf28}+0.62\%$
test_dqn_speed[reduce-overhead-None] 0.7062ms 0.5743ms 1.7412 KOps/s 1.7024 KOps/s $\color{#35bf28}+2.28\%$
test_dqn_speed[reduce-overhead-backward] 1.1158ms 1.0672ms 936.9911 Ops/s 928.9049 Ops/s $\color{#35bf28}+0.87\%$
test_ddpg_speed[False-None] 3.2560ms 2.9042ms 344.3250 Ops/s 345.2099 Ops/s $\color{#d91a1a}-0.26\%$
test_ddpg_speed[False-backward] 4.6526ms 4.2895ms 233.1258 Ops/s 236.4968 Ops/s $\color{#d91a1a}-1.43\%$
test_ddpg_speed[True-None] 1.5100ms 1.3522ms 739.5176 Ops/s 726.8062 Ops/s $\color{#35bf28}+1.75\%$
test_ddpg_speed[True-backward] 2.7091ms 2.5814ms 387.3905 Ops/s 379.1427 Ops/s $\color{#35bf28}+2.18\%$
test_ddpg_speed[reduce-overhead-None] 1.5034ms 1.3583ms 736.2068 Ops/s 732.6062 Ops/s $\color{#35bf28}+0.49\%$
test_ddpg_speed[reduce-overhead-backward] 2.0939ms 2.0431ms 489.4478 Ops/s 487.7401 Ops/s $\color{#35bf28}+0.35\%$
test_sac_speed[False-None] 8.5939ms 8.1495ms 122.7063 Ops/s 120.3677 Ops/s $\color{#35bf28}+1.94\%$
test_sac_speed[False-backward] 11.8218ms 11.3004ms 88.4926 Ops/s 88.8688 Ops/s $\color{#d91a1a}-0.42\%$
test_sac_speed[True-None] 1.9113ms 1.8542ms 539.3257 Ops/s 515.5893 Ops/s $\color{#35bf28}+4.60\%$
test_sac_speed[True-backward] 3.8092ms 3.7191ms 268.8830 Ops/s 277.1748 Ops/s $\color{#d91a1a}-2.99\%$
test_sac_speed[reduce-overhead-None] 21.8164ms 12.0707ms 82.8453 Ops/s 82.3708 Ops/s $\color{#35bf28}+0.58\%$
test_sac_speed[reduce-overhead-backward] 1.8455ms 1.7907ms 558.4284 Ops/s 598.5687 Ops/s $\textbf{\color{#d91a1a}-6.71\%}$
test_redq_speed[False-None] 7.9858ms 7.5972ms 131.6281 Ops/s 132.4608 Ops/s $\color{#d91a1a}-0.63\%$
test_redq_speed[False-backward] 12.4912ms 11.8514ms 84.3780 Ops/s 87.0358 Ops/s $\color{#d91a1a}-3.05\%$
test_redq_speed[True-None] 2.3477ms 2.2926ms 436.1868 Ops/s 429.5329 Ops/s $\color{#35bf28}+1.55\%$
test_redq_speed[True-backward] 4.3778ms 4.2108ms 237.4831 Ops/s 245.2513 Ops/s $\color{#d91a1a}-3.17\%$
test_redq_speed[reduce-overhead-None] 2.3736ms 2.3057ms 433.7151 Ops/s 421.9012 Ops/s $\color{#35bf28}+2.80\%$
test_redq_speed[reduce-overhead-backward] 4.3571ms 4.1856ms 238.9170 Ops/s 244.6957 Ops/s $\color{#d91a1a}-2.36\%$
test_redq_deprec_speed[False-None] 9.4403ms 9.1047ms 109.8334 Ops/s 109.7965 Ops/s $\color{#35bf28}+0.03\%$
test_redq_deprec_speed[False-backward] 12.6247ms 12.2409ms 81.6936 Ops/s 82.9340 Ops/s $\color{#d91a1a}-1.50\%$
test_redq_deprec_speed[True-None] 2.8149ms 2.6453ms 378.0276 Ops/s 376.1807 Ops/s $\color{#35bf28}+0.49\%$
test_redq_deprec_speed[True-backward] 5.4264ms 4.6303ms 215.9692 Ops/s 220.5850 Ops/s $\color{#d91a1a}-2.09\%$
test_redq_deprec_speed[reduce-overhead-None] 3.2108ms 2.7073ms 369.3771 Ops/s 374.2367 Ops/s $\color{#d91a1a}-1.30\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.5343ms 4.4740ms 223.5126 Ops/s 222.7144 Ops/s $\color{#35bf28}+0.36\%$
test_td3_speed[False-None] 8.1965ms 8.0354ms 124.4499 Ops/s 124.4923 Ops/s $\color{#d91a1a}-0.03\%$
test_td3_speed[False-backward] 11.0572ms 10.5798ms 94.5198 Ops/s 96.7360 Ops/s $\color{#d91a1a}-2.29\%$
test_td3_speed[True-None] 1.6751ms 1.6441ms 608.2356 Ops/s 584.4582 Ops/s $\color{#35bf28}+4.07\%$
test_td3_speed[True-backward] 3.5030ms 3.3531ms 298.2352 Ops/s 290.7556 Ops/s $\color{#35bf28}+2.57\%$
test_td3_speed[reduce-overhead-None] 54.8140ms 26.6419ms 37.5349 Ops/s 36.1504 Ops/s $\color{#35bf28}+3.83\%$
test_td3_speed[reduce-overhead-backward] 1.5634ms 1.5087ms 662.8190 Ops/s 713.3949 Ops/s $\textbf{\color{#d91a1a}-7.09\%}$
test_cql_speed[False-None] 17.3755ms 16.9174ms 59.1109 Ops/s 59.1351 Ops/s $\color{#d91a1a}-0.04\%$
test_cql_speed[False-backward] 22.8785ms 22.4282ms 44.5868 Ops/s 45.2609 Ops/s $\color{#d91a1a}-1.49\%$
test_cql_speed[True-None] 3.4442ms 3.2691ms 305.8985 Ops/s 303.1266 Ops/s $\color{#35bf28}+0.91\%$
test_cql_speed[True-backward] 5.9072ms 5.5250ms 180.9965 Ops/s 174.9584 Ops/s $\color{#35bf28}+3.45\%$
test_cql_speed[reduce-overhead-None] 21.6988ms 13.3295ms 75.0214 Ops/s 57.7355 Ops/s $\textbf{\color{#35bf28}+29.94\%}$
test_cql_speed[reduce-overhead-backward] 1.9756ms 1.8471ms 541.3949 Ops/s 491.7953 Ops/s $\textbf{\color{#35bf28}+10.09\%}$
test_a2c_speed[False-None] 3.6608ms 3.2046ms 312.0490 Ops/s 308.2761 Ops/s $\color{#35bf28}+1.22\%$
test_a2c_speed[False-backward] 6.7943ms 6.0774ms 164.5428 Ops/s 158.5108 Ops/s $\color{#35bf28}+3.81\%$
test_a2c_speed[True-None] 1.7935ms 1.3516ms 739.8846 Ops/s 738.5389 Ops/s $\color{#35bf28}+0.18\%$
test_a2c_speed[True-backward] 3.0625ms 2.8989ms 344.9545 Ops/s 336.8912 Ops/s $\color{#35bf28}+2.39\%$
test_a2c_speed[reduce-overhead-None] 16.1986ms 9.1413ms 109.3941 Ops/s 110.9867 Ops/s $\color{#d91a1a}-1.43\%$
test_a2c_speed[reduce-overhead-backward] 1.6111ms 1.4722ms 679.2545 Ops/s 673.7948 Ops/s $\color{#35bf28}+0.81\%$
test_ppo_speed[False-None] 3.9991ms 3.7336ms 267.8376 Ops/s 271.2877 Ops/s $\color{#d91a1a}-1.27\%$
test_ppo_speed[False-backward] 7.4877ms 6.9015ms 144.8957 Ops/s 147.6619 Ops/s $\color{#d91a1a}-1.87\%$
test_ppo_speed[True-None] 1.8251ms 1.4102ms 709.1168 Ops/s 702.1935 Ops/s $\color{#35bf28}+0.99\%$
test_ppo_speed[True-backward] 3.4598ms 3.0738ms 325.3265 Ops/s 320.5326 Ops/s $\color{#35bf28}+1.50\%$
test_ppo_speed[reduce-overhead-None] 1.1833ms 0.9730ms 1.0278 KOps/s 1.0213 KOps/s $\color{#35bf28}+0.64\%$
test_ppo_speed[reduce-overhead-backward] 1.5474ms 1.4041ms 712.2091 Ops/s 681.7440 Ops/s $\color{#35bf28}+4.47\%$
test_reinforce_speed[False-None] 2.4960ms 2.2976ms 435.2342 Ops/s 436.6165 Ops/s $\color{#d91a1a}-0.32\%$
test_reinforce_speed[False-backward] 3.3327ms 3.2814ms 304.7467 Ops/s 303.1195 Ops/s $\color{#35bf28}+0.54\%$
test_reinforce_speed[True-None] 1.4496ms 1.2957ms 771.7699 Ops/s 756.0658 Ops/s $\color{#35bf28}+2.08\%$
test_reinforce_speed[True-backward] 3.0987ms 2.9445ms 339.6138 Ops/s 340.2036 Ops/s $\color{#d91a1a}-0.17\%$
test_reinforce_speed[reduce-overhead-None] 21.0431ms 10.1217ms 98.7977 Ops/s 99.6718 Ops/s $\color{#d91a1a}-0.88\%$
test_reinforce_speed[reduce-overhead-backward] 1.5674ms 1.4776ms 676.7918 Ops/s 658.4939 Ops/s $\color{#35bf28}+2.78\%$
test_iql_speed[False-None] 9.6786ms 9.2128ms 108.5446 Ops/s 107.7558 Ops/s $\color{#35bf28}+0.73\%$
test_iql_speed[False-backward] 13.4050ms 12.8554ms 77.7886 Ops/s 76.8478 Ops/s $\color{#35bf28}+1.22\%$
test_iql_speed[True-None] 2.4157ms 2.2420ms 446.0354 Ops/s 435.3140 Ops/s $\color{#35bf28}+2.46\%$
test_iql_speed[True-backward] 4.9614ms 4.7704ms 209.6266 Ops/s 195.7140 Ops/s $\textbf{\color{#35bf28}+7.11\%}$
test_iql_speed[reduce-overhead-None] 18.7979ms 11.1807ms 89.4397 Ops/s 89.1122 Ops/s $\color{#35bf28}+0.37\%$
test_iql_speed[reduce-overhead-backward] 2.0540ms 1.9055ms 524.7945 Ops/s 461.0686 Ops/s $\textbf{\color{#35bf28}+13.82\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9157ms 6.3075ms 158.5402 Ops/s 155.2248 Ops/s $\color{#35bf28}+2.14\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5491ms 0.3557ms 2.8114 KOps/s 2.7532 KOps/s $\color{#35bf28}+2.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7587ms 0.3348ms 2.9869 KOps/s 3.2612 KOps/s $\textbf{\color{#d91a1a}-8.41\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.6281ms 6.0723ms 164.6836 Ops/s 164.1799 Ops/s $\color{#35bf28}+0.31\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1575ms 0.3610ms 2.7700 KOps/s 2.9346 KOps/s $\textbf{\color{#d91a1a}-5.61\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5575ms 0.3209ms 3.1160 KOps/s 2.8068 KOps/s $\textbf{\color{#35bf28}+11.02\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7528ms 1.4411ms 693.8904 Ops/s 701.6493 Ops/s $\color{#d91a1a}-1.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4498ms 1.2054ms 829.6321 Ops/s 768.9084 Ops/s $\textbf{\color{#35bf28}+7.90\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4845ms 6.2512ms 159.9701 Ops/s 158.4629 Ops/s $\color{#35bf28}+0.95\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9070ms 0.4346ms 2.3011 KOps/s 2.3415 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6186ms 0.3991ms 2.5057 KOps/s 2.5344 KOps/s $\color{#d91a1a}-1.13\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.5033ms 6.1321ms 163.0764 Ops/s 162.1462 Ops/s $\color{#35bf28}+0.57\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9780ms 0.3524ms 2.8378 KOps/s 2.8852 KOps/s $\color{#d91a1a}-1.64\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6366ms 0.3247ms 3.0799 KOps/s 3.9762 KOps/s $\textbf{\color{#d91a1a}-22.54\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.8704ms 6.1194ms 163.4146 Ops/s 164.1931 Ops/s $\color{#d91a1a}-0.47\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0325ms 0.3517ms 2.8434 KOps/s 3.4356 KOps/s $\textbf{\color{#d91a1a}-17.24\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5976ms 0.3262ms 3.0659 KOps/s 3.7479 KOps/s $\textbf{\color{#d91a1a}-18.20\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4377ms 6.2018ms 161.2427 Ops/s 157.0630 Ops/s $\color{#35bf28}+2.66\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0604ms 0.4192ms 2.3857 KOps/s 1.9352 KOps/s $\textbf{\color{#35bf28}+23.28\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6264ms 0.4695ms 2.1299 KOps/s 2.0000 KOps/s $\textbf{\color{#35bf28}+6.50\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1225ms 5.5400ms 180.5063 Ops/s 174.0703 Ops/s $\color{#35bf28}+3.70\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 11.1219ms 2.1229ms 471.0555 Ops/s 434.4218 Ops/s $\textbf{\color{#35bf28}+8.43\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.1263ms 1.1330ms 882.5755 Ops/s 829.5414 Ops/s $\textbf{\color{#35bf28}+6.39\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 8.8680ms 5.6875ms 175.8232 Ops/s 176.5598 Ops/s $\color{#d91a1a}-0.42\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 6.2524ms 2.0133ms 496.7064 Ops/s 427.5214 Ops/s $\textbf{\color{#35bf28}+16.18\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.6319ms 1.3066ms 765.3741 Ops/s 848.3594 Ops/s $\textbf{\color{#d91a1a}-9.78\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4950s 15.6163ms 64.0356 Ops/s 31.0911 Ops/s $\textbf{\color{#35bf28}+105.96\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.3876ms 2.2349ms 447.4468 Ops/s 452.2604 Ops/s $\color{#d91a1a}-1.06\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.8064ms 1.4187ms 704.8853 Ops/s 680.9249 Ops/s $\color{#35bf28}+3.52\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.8750ms 13.0574ms 76.5849 Ops/s 70.7037 Ops/s $\textbf{\color{#35bf28}+8.32\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.3288ms 16.8073ms 59.4980 Ops/s 53.4379 Ops/s $\textbf{\color{#35bf28}+11.34\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 17.8810ms 17.6887ms 56.5332 Ops/s 55.4348 Ops/s $\color{#35bf28}+1.98\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7483ms 17.1778ms 58.2148 Ops/s 57.9630 Ops/s $\color{#35bf28}+0.43\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.2747ms 17.6515ms 56.6524 Ops/s 54.7895 Ops/s $\color{#35bf28}+3.40\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.0295ms 18.4197ms 54.2896 Ops/s 53.2645 Ops/s $\color{#35bf28}+1.92\%$

[ghstack-poisoned]
@vmoens vmoens added bc breaking backward compatibility breaking change Deprecation labels Feb 4, 2025
@vmoens vmoens merged commit 9fcdb11 into gh/vmoens/91/base Feb 4, 2025
29 of 51 checks passed
vmoens added a commit that referenced this pull request Feb 4, 2025
ghstack-source-id: 78f248e1239a04fc5213aa4418a158f741679593
Pull Request resolved: #2748
@vmoens vmoens deleted the gh/vmoens/91/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