Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 28, 2025

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 28, 2025
ghstack-source-id: c625cd3
Pull-Request: #3225
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 28, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3225

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.

@meta-cla meta-cla 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 Oct 28, 2025
@vmoens vmoens added the bug Something isn't working label Oct 28, 2025
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Nov 6, 2025

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}25$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 81.8127μs 80.8809μs 12.3639 KOps/s 12.2836 KOps/s $\color{#35bf28}+0.65\%$
test_tensor_to_bytestream_speed[torch.save] 0.1415ms 0.1391ms 7.1867 KOps/s 7.2797 KOps/s $\color{#d91a1a}-1.28\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1119s 0.1118s 8.9462 Ops/s 9.1229 Ops/s $\color{#d91a1a}-1.94\%$
test_tensor_to_bytestream_speed[numpy] 2.8446μs 2.8310μs 353.2374 KOps/s 370.2748 KOps/s $\color{#d91a1a}-4.60\%$
test_tensor_to_bytestream_speed[safetensors] 43.1634μs 42.0536μs 23.7792 KOps/s 23.4233 KOps/s $\color{#35bf28}+1.52\%$
test_simple 0.6479s 0.5546s 1.8032 Ops/s 1.8037 Ops/s $\color{#d91a1a}-0.03\%$
test_transformed 1.1946s 1.1027s 0.9069 Ops/s 0.9070 Ops/s $\color{#d91a1a}-0.01\%$
test_serial 1.6025s 1.5941s 0.6273 Ops/s 0.6149 Ops/s $\color{#35bf28}+2.03\%$
test_parallel 1.0379s 1.0182s 0.9822 Ops/s 0.9429 Ops/s $\color{#35bf28}+4.17\%$
test_step_mdp_speed[True-True-True-True-True] 83.0010μs 42.4922μs 23.5337 KOps/s 23.0879 KOps/s $\color{#35bf28}+1.93\%$
test_step_mdp_speed[True-True-True-True-False] 46.4410μs 23.7269μs 42.1462 KOps/s 40.7423 KOps/s $\color{#35bf28}+3.45\%$
test_step_mdp_speed[True-True-True-False-True] 55.1320μs 23.8880μs 41.8620 KOps/s 40.5281 KOps/s $\color{#35bf28}+3.29\%$
test_step_mdp_speed[True-True-True-False-False] 40.6610μs 13.3207μs 75.0711 KOps/s 74.0592 KOps/s $\color{#35bf28}+1.37\%$
test_step_mdp_speed[True-True-False-True-True] 0.1038ms 46.6584μs 21.4324 KOps/s 21.1558 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[True-True-False-True-False] 52.1110μs 26.6790μs 37.4826 KOps/s 36.4918 KOps/s $\color{#35bf28}+2.72\%$
test_step_mdp_speed[True-True-False-False-True] 57.2610μs 26.6021μs 37.5910 KOps/s 35.8725 KOps/s $\color{#35bf28}+4.79\%$
test_step_mdp_speed[True-True-False-False-False] 49.5210μs 15.7555μs 63.4697 KOps/s 60.9065 KOps/s $\color{#35bf28}+4.21\%$
test_step_mdp_speed[True-False-True-True-True] 0.1202ms 48.7679μs 20.5053 KOps/s 20.2025 KOps/s $\color{#35bf28}+1.50\%$
test_step_mdp_speed[True-False-True-True-False] 70.4020μs 29.0509μs 34.4223 KOps/s 33.3304 KOps/s $\color{#35bf28}+3.28\%$
test_step_mdp_speed[True-False-True-False-True] 58.8010μs 26.2971μs 38.0270 KOps/s 36.6745 KOps/s $\color{#35bf28}+3.69\%$
test_step_mdp_speed[True-False-True-False-False] 44.9110μs 15.6272μs 63.9908 KOps/s 61.6961 KOps/s $\color{#35bf28}+3.72\%$
test_step_mdp_speed[True-False-False-True-True] 94.3520μs 50.3146μs 19.8749 KOps/s 19.4089 KOps/s $\color{#35bf28}+2.40\%$
test_step_mdp_speed[True-False-False-True-False] 97.9320μs 31.4120μs 31.8350 KOps/s 30.8143 KOps/s $\color{#35bf28}+3.31\%$
test_step_mdp_speed[True-False-False-False-True] 61.9210μs 28.8156μs 34.7034 KOps/s 33.9459 KOps/s $\color{#35bf28}+2.23\%$
test_step_mdp_speed[True-False-False-False-False] 0.1469ms 18.2820μs 54.6985 KOps/s 53.3457 KOps/s $\color{#35bf28}+2.54\%$
test_step_mdp_speed[False-True-True-True-True] 81.9220μs 48.5491μs 20.5977 KOps/s 19.9886 KOps/s $\color{#35bf28}+3.05\%$
test_step_mdp_speed[False-True-True-True-False] 57.1410μs 29.3043μs 34.1247 KOps/s 33.6583 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[False-True-True-False-True] 2.8241ms 30.4157μs 32.8778 KOps/s 31.7549 KOps/s $\color{#35bf28}+3.54\%$
test_step_mdp_speed[False-True-True-False-False] 51.8910μs 17.6644μs 56.6110 KOps/s 55.9871 KOps/s $\color{#35bf28}+1.11\%$
test_step_mdp_speed[False-True-False-True-True] 0.1053ms 51.9533μs 19.2481 KOps/s 19.1851 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[False-True-False-True-False] 60.3010μs 31.2380μs 32.0123 KOps/s 30.3073 KOps/s $\textbf{\color{#35bf28}+5.63\%}$
test_step_mdp_speed[False-True-False-False-True] 0.2185ms 32.1391μs 31.1147 KOps/s 29.7702 KOps/s $\color{#35bf28}+4.52\%$
test_step_mdp_speed[False-True-False-False-False] 51.7010μs 19.9989μs 50.0027 KOps/s 47.7334 KOps/s $\color{#35bf28}+4.75\%$
test_step_mdp_speed[False-False-True-True-True] 0.1134ms 53.1397μs 18.8183 KOps/s 17.9958 KOps/s $\color{#35bf28}+4.57\%$
test_step_mdp_speed[False-False-True-True-False] 81.1920μs 34.1764μs 29.2599 KOps/s 28.4026 KOps/s $\color{#35bf28}+3.02\%$
test_step_mdp_speed[False-False-True-False-True] 0.2002ms 33.0299μs 30.2756 KOps/s 29.9780 KOps/s $\color{#35bf28}+0.99\%$
test_step_mdp_speed[False-False-True-False-False] 65.6910μs 20.1817μs 49.5499 KOps/s 48.8090 KOps/s $\color{#35bf28}+1.52\%$
test_step_mdp_speed[False-False-False-True-True] 0.1065ms 55.6390μs 17.9730 KOps/s 17.8704 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[False-False-False-True-False] 74.3610μs 37.5752μs 26.6133 KOps/s 26.4000 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[False-False-False-False-True] 0.1050ms 35.2731μs 28.3502 KOps/s 28.1183 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-False-False-False-False] 72.3920μs 22.9958μs 43.4862 KOps/s 44.0330 KOps/s $\color{#d91a1a}-1.24\%$
test_values[generalized_advantage_estimate-True-True] 10.2369ms 9.4189ms 106.1699 Ops/s 109.2326 Ops/s $\color{#d91a1a}-2.80\%$
test_values[vec_generalized_advantage_estimate-True-True] 13.7984ms 11.4152ms 87.6023 Ops/s 86.5328 Ops/s $\color{#35bf28}+1.24\%$
test_values[td0_return_estimate-False-False] 0.2532ms 0.1314ms 7.6123 KOps/s 7.8013 KOps/s $\color{#d91a1a}-2.42\%$
test_values[td1_return_estimate-False-False] 26.3995ms 25.8614ms 38.6677 Ops/s 39.6542 Ops/s $\color{#d91a1a}-2.49\%$
test_values[vec_td1_return_estimate-False-False] 11.9544ms 11.5525ms 86.5611 Ops/s 85.9921 Ops/s $\color{#35bf28}+0.66\%$
test_values[td_lambda_return_estimate-True-False] 40.0303ms 38.9420ms 25.6792 Ops/s 26.5726 Ops/s $\color{#d91a1a}-3.36\%$
test_values[vec_td_lambda_return_estimate-True-False] 14.8865ms 12.1794ms 82.1060 Ops/s 86.2883 Ops/s $\color{#d91a1a}-4.85\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.2592ms 8.1595ms 122.5566 Ops/s 126.9789 Ops/s $\color{#d91a1a}-3.48\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.5919ms 1.4480ms 690.6018 Ops/s 645.5540 Ops/s $\textbf{\color{#35bf28}+6.98\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5313ms 0.3851ms 2.5967 KOps/s 2.5837 KOps/s $\color{#35bf28}+0.50\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 25.5997ms 24.3528ms 41.0631 Ops/s 33.0727 Ops/s $\textbf{\color{#35bf28}+24.16\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8607ms 1.7069ms 585.8544 Ops/s 582.6469 Ops/s $\color{#35bf28}+0.55\%$
test_dqn_speed[False-None] 6.6755ms 1.4050ms 711.7489 Ops/s 721.9376 Ops/s $\color{#d91a1a}-1.41\%$
test_dqn_speed[False-backward] 1.9490ms 1.9031ms 525.4574 Ops/s 533.6631 Ops/s $\color{#d91a1a}-1.54\%$
test_dqn_speed[True-None] 0.9273ms 0.5293ms 1.8894 KOps/s 1.9028 KOps/s $\color{#d91a1a}-0.70\%$
test_dqn_speed[True-backward] 1.0009ms 0.9610ms 1.0406 KOps/s 876.0030 Ops/s $\textbf{\color{#35bf28}+18.79\%}$
test_dqn_speed[reduce-overhead-None] 0.8003ms 0.5179ms 1.9309 KOps/s 1.9167 KOps/s $\color{#35bf28}+0.74\%$
test_dqn_speed[reduce-overhead-backward] 0.9855ms 0.9475ms 1.0554 KOps/s 892.1248 Ops/s $\textbf{\color{#35bf28}+18.30\%}$
test_ddpg_speed[False-None] 3.1657ms 2.7893ms 358.5073 Ops/s 350.2572 Ops/s $\color{#35bf28}+2.36\%$
test_ddpg_speed[False-backward] 4.3710ms 3.9834ms 251.0441 Ops/s 246.8627 Ops/s $\color{#35bf28}+1.69\%$
test_ddpg_speed[True-None] 1.7598ms 1.3816ms 723.8024 Ops/s 733.3311 Ops/s $\color{#d91a1a}-1.30\%$
test_ddpg_speed[True-backward] 2.3789ms 2.3219ms 430.6782 Ops/s 352.0748 Ops/s $\textbf{\color{#35bf28}+22.33\%}$
test_ddpg_speed[reduce-overhead-None] 1.7101ms 1.3682ms 730.9089 Ops/s 731.4963 Ops/s $\color{#d91a1a}-0.08\%$
test_ddpg_speed[reduce-overhead-backward] 2.4666ms 2.3478ms 425.9247 Ops/s 430.8815 Ops/s $\color{#d91a1a}-1.15\%$
test_sac_speed[False-None] 8.1995ms 7.5869ms 131.8059 Ops/s 131.1705 Ops/s $\color{#35bf28}+0.48\%$
test_sac_speed[False-backward] 11.1381ms 10.8203ms 92.4193 Ops/s 92.6576 Ops/s $\color{#d91a1a}-0.26\%$
test_sac_speed[True-None] 2.5334ms 2.1272ms 470.1003 Ops/s 459.3650 Ops/s $\color{#35bf28}+2.34\%$
test_sac_speed[True-backward] 4.2581ms 4.0469ms 247.1025 Ops/s 249.5379 Ops/s $\color{#d91a1a}-0.98\%$
test_sac_speed[reduce-overhead-None] 2.5494ms 2.1350ms 468.3835 Ops/s 471.1503 Ops/s $\color{#d91a1a}-0.59\%$
test_sac_speed[reduce-overhead-backward] 4.1852ms 4.0341ms 247.8896 Ops/s 247.0699 Ops/s $\color{#35bf28}+0.33\%$
test_redq_speed[False-None] 10.9822ms 10.4819ms 95.4021 Ops/s 95.1399 Ops/s $\color{#35bf28}+0.28\%$
test_redq_speed[False-backward] 19.3302ms 18.4490ms 54.2036 Ops/s 53.7718 Ops/s $\color{#35bf28}+0.80\%$
test_redq_speed[True-None] 4.4890ms 4.1920ms 238.5519 Ops/s 211.7610 Ops/s $\textbf{\color{#35bf28}+12.65\%}$
test_redq_speed[True-backward] 10.1217ms 9.8144ms 101.8913 Ops/s 99.3313 Ops/s $\color{#35bf28}+2.58\%$
test_redq_speed[reduce-overhead-None] 4.6282ms 4.3632ms 229.1897 Ops/s 236.6738 Ops/s $\color{#d91a1a}-3.16\%$
test_redq_speed[reduce-overhead-backward] 10.5269ms 10.0543ms 99.4597 Ops/s 98.2126 Ops/s $\color{#35bf28}+1.27\%$
test_redq_deprec_speed[False-None] 11.6326ms 10.9699ms 91.1588 Ops/s 91.8642 Ops/s $\color{#d91a1a}-0.77\%$
test_redq_deprec_speed[False-backward] 16.8385ms 16.1632ms 61.8691 Ops/s 64.2886 Ops/s $\color{#d91a1a}-3.76\%$
test_redq_deprec_speed[True-None] 3.9078ms 3.7030ms 270.0491 Ops/s 283.7977 Ops/s $\color{#d91a1a}-4.84\%$
test_redq_deprec_speed[True-backward] 8.2536ms 7.8438ms 127.4886 Ops/s 122.7781 Ops/s $\color{#35bf28}+3.84\%$
test_redq_deprec_speed[reduce-overhead-None] 4.1874ms 3.6472ms 274.1851 Ops/s 277.9023 Ops/s $\color{#d91a1a}-1.34\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.5199ms 7.9510ms 125.7701 Ops/s 116.8335 Ops/s $\textbf{\color{#35bf28}+7.65\%}$
test_td3_speed[False-None] 7.7986ms 7.6313ms 131.0401 Ops/s 130.4513 Ops/s $\color{#35bf28}+0.45\%$
test_td3_speed[False-backward] 11.1316ms 10.4181ms 95.9871 Ops/s 96.0128 Ops/s $\color{#d91a1a}-0.03\%$
test_td3_speed[True-None] 1.8650ms 1.8242ms 548.1793 Ops/s 556.6682 Ops/s $\color{#d91a1a}-1.52\%$
test_td3_speed[True-backward] 3.6883ms 3.5434ms 282.2159 Ops/s 285.3815 Ops/s $\color{#d91a1a}-1.11\%$
test_td3_speed[reduce-overhead-None] 1.8014ms 1.7585ms 568.6595 Ops/s 576.8067 Ops/s $\color{#d91a1a}-1.41\%$
test_td3_speed[reduce-overhead-backward] 3.7895ms 3.5453ms 282.0670 Ops/s 283.3835 Ops/s $\color{#d91a1a}-0.46\%$
test_cql_speed[False-None] 28.1065ms 25.4688ms 39.2638 Ops/s 39.9612 Ops/s $\color{#d91a1a}-1.75\%$
test_cql_speed[False-backward] 39.4540ms 35.0949ms 28.4942 Ops/s 28.7721 Ops/s $\color{#d91a1a}-0.97\%$
test_cql_speed[True-None] 12.6013ms 12.1517ms 82.2931 Ops/s 81.7070 Ops/s $\color{#35bf28}+0.72\%$
test_cql_speed[True-backward] 18.0334ms 17.7754ms 56.2576 Ops/s 57.0563 Ops/s $\color{#d91a1a}-1.40\%$
test_cql_speed[reduce-overhead-None] 12.6591ms 12.2682ms 81.5116 Ops/s 68.6981 Ops/s $\textbf{\color{#35bf28}+18.65\%}$
test_cql_speed[reduce-overhead-backward] 18.4568ms 17.9197ms 55.8044 Ops/s 52.4636 Ops/s $\textbf{\color{#35bf28}+6.37\%}$
test_a2c_speed[False-None] 5.6797ms 5.3164ms 188.0958 Ops/s 181.4974 Ops/s $\color{#35bf28}+3.64\%$
test_a2c_speed[False-backward] 12.0496ms 11.7422ms 85.1629 Ops/s 80.8707 Ops/s $\textbf{\color{#35bf28}+5.31\%}$
test_a2c_speed[True-None] 4.0659ms 3.6096ms 277.0356 Ops/s 258.9165 Ops/s $\textbf{\color{#35bf28}+7.00\%}$
test_a2c_speed[True-backward] 9.2681ms 8.6104ms 116.1391 Ops/s 110.6938 Ops/s $\color{#35bf28}+4.92\%$
test_a2c_speed[reduce-overhead-None] 3.9952ms 3.6655ms 272.8130 Ops/s 267.1881 Ops/s $\color{#35bf28}+2.11\%$
test_a2c_speed[reduce-overhead-backward] 9.3370ms 8.9220ms 112.0820 Ops/s 112.2445 Ops/s $\color{#d91a1a}-0.14\%$
test_ppo_speed[False-None] 6.2441ms 5.9028ms 169.4098 Ops/s 172.8086 Ops/s $\color{#d91a1a}-1.97\%$
test_ppo_speed[False-backward] 13.1830ms 12.7506ms 78.4278 Ops/s 80.7412 Ops/s $\color{#d91a1a}-2.87\%$
test_ppo_speed[True-None] 3.9764ms 3.6849ms 271.3755 Ops/s 262.1833 Ops/s $\color{#35bf28}+3.51\%$
test_ppo_speed[True-backward] 8.8961ms 8.5736ms 116.6368 Ops/s 119.9286 Ops/s $\color{#d91a1a}-2.74\%$
test_ppo_speed[reduce-overhead-None] 4.0210ms 3.6614ms 273.1204 Ops/s 276.3406 Ops/s $\color{#d91a1a}-1.17\%$
test_ppo_speed[reduce-overhead-backward] 9.5587ms 9.0517ms 110.4767 Ops/s 112.9264 Ops/s $\color{#d91a1a}-2.17\%$
test_reinforce_speed[False-None] 4.9744ms 4.6730ms 213.9942 Ops/s 220.5011 Ops/s $\color{#d91a1a}-2.95\%$
test_reinforce_speed[False-backward] 8.0239ms 7.5977ms 131.6184 Ops/s 138.7940 Ops/s $\textbf{\color{#d91a1a}-5.17\%}$
test_reinforce_speed[True-None] 3.1799ms 2.9185ms 342.6441 Ops/s 346.1066 Ops/s $\color{#d91a1a}-1.00\%$
test_reinforce_speed[True-backward] 8.2101ms 7.9335ms 126.0473 Ops/s 123.8756 Ops/s $\color{#35bf28}+1.75\%$
test_reinforce_speed[reduce-overhead-None] 3.1905ms 2.8786ms 347.3940 Ops/s 358.2828 Ops/s $\color{#d91a1a}-3.04\%$
test_reinforce_speed[reduce-overhead-backward] 8.5452ms 8.2509ms 121.1995 Ops/s 122.1810 Ops/s $\color{#d91a1a}-0.80\%$
test_iql_speed[False-None] 26.7048ms 21.2376ms 47.0862 Ops/s 50.4251 Ops/s $\textbf{\color{#d91a1a}-6.62\%}$
test_iql_speed[False-backward] 37.4494ms 31.4597ms 31.7867 Ops/s 33.3887 Ops/s $\color{#d91a1a}-4.80\%$
test_iql_speed[True-None] 10.9234ms 9.0496ms 110.5022 Ops/s 118.4375 Ops/s $\textbf{\color{#d91a1a}-6.70\%}$
test_iql_speed[True-backward] 17.5751ms 16.8685ms 59.2821 Ops/s 59.8947 Ops/s $\color{#d91a1a}-1.02\%$
test_iql_speed[reduce-overhead-None] 8.8339ms 8.4188ms 118.7824 Ops/s 116.1695 Ops/s $\color{#35bf28}+2.25\%$
test_iql_speed[reduce-overhead-backward] 16.9852ms 16.6190ms 60.1720 Ops/s 52.3015 Ops/s $\textbf{\color{#35bf28}+15.05\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.2531ms 5.7932ms 172.6158 Ops/s 173.0147 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6674ms 0.2918ms 3.4265 KOps/s 2.7183 KOps/s $\textbf{\color{#35bf28}+26.05\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5921ms 0.2916ms 3.4292 KOps/s 2.8331 KOps/s $\textbf{\color{#35bf28}+21.04\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7538ms 5.4976ms 181.8967 Ops/s 180.1857 Ops/s $\color{#35bf28}+0.95\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8091ms 0.2853ms 3.5051 KOps/s 2.7930 KOps/s $\textbf{\color{#35bf28}+25.50\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5085ms 0.2648ms 3.7759 KOps/s 2.9488 KOps/s $\textbf{\color{#35bf28}+28.05\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6192ms 1.2020ms 831.9440 Ops/s 744.5361 Ops/s $\textbf{\color{#35bf28}+11.74\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.2885ms 1.1141ms 897.6078 Ops/s 786.2799 Ops/s $\textbf{\color{#35bf28}+14.16\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8055ms 5.6393ms 177.3256 Ops/s 174.7009 Ops/s $\color{#35bf28}+1.50\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8939ms 0.4510ms 2.2172 KOps/s 1.8470 KOps/s $\textbf{\color{#35bf28}+20.04\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7652ms 0.4249ms 2.3537 KOps/s 2.4454 KOps/s $\color{#d91a1a}-3.75\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.7733ms 5.5317ms 180.7774 Ops/s 177.5953 Ops/s $\color{#35bf28}+1.79\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8308ms 0.2736ms 3.6554 KOps/s 3.1161 KOps/s $\textbf{\color{#35bf28}+17.31\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5636ms 0.2642ms 3.7851 KOps/s 2.9123 KOps/s $\textbf{\color{#35bf28}+29.97\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7381ms 5.4949ms 181.9855 Ops/s 179.4367 Ops/s $\color{#35bf28}+1.42\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7510ms 0.3351ms 2.9846 KOps/s 3.6094 KOps/s $\textbf{\color{#d91a1a}-17.31\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5819ms 0.3254ms 3.0735 KOps/s 3.6124 KOps/s $\textbf{\color{#d91a1a}-14.92\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8043ms 5.6774ms 176.1363 Ops/s 172.8817 Ops/s $\color{#35bf28}+1.88\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2873ms 0.5232ms 1.9112 KOps/s 2.0196 KOps/s $\textbf{\color{#d91a1a}-5.37\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9768ms 0.4599ms 2.1744 KOps/s 1.9965 KOps/s $\textbf{\color{#35bf28}+8.91\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.4756s 14.2884ms 69.9867 Ops/s 199.2154 Ops/s $\textbf{\color{#d91a1a}-64.87\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.0668ms 1.6495ms 606.2480 Ops/s 492.7956 Ops/s $\textbf{\color{#35bf28}+23.02\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.9489ms 1.1627ms 860.0482 Ops/s 819.1576 Ops/s $\color{#35bf28}+4.99\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.7784ms 4.9067ms 203.8034 Ops/s 198.1014 Ops/s $\color{#35bf28}+2.88\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.5102ms 1.9787ms 505.3924 Ops/s 494.1460 Ops/s $\color{#35bf28}+2.28\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.0921ms 1.1573ms 864.0739 Ops/s 837.4760 Ops/s $\color{#35bf28}+3.18\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4261s 13.5322ms 73.8978 Ops/s 58.4152 Ops/s $\textbf{\color{#35bf28}+26.50\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 12.5964ms 2.0652ms 484.2082 Ops/s 483.1394 Ops/s $\color{#35bf28}+0.22\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.3618ms 0.9880ms 1.0122 KOps/s 990.8235 Ops/s $\color{#35bf28}+2.15\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 34.7372ms 32.2197ms 31.0369 Ops/s 30.7911 Ops/s $\color{#35bf28}+0.80\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.8315ms 16.8651ms 59.2940 Ops/s 57.2944 Ops/s $\color{#35bf28}+3.49\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 34.6712ms 32.7790ms 30.5073 Ops/s 29.7257 Ops/s $\color{#35bf28}+2.63\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7779ms 16.9928ms 58.8484 Ops/s 56.5246 Ops/s $\color{#35bf28}+4.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 0.3950s 41.5206ms 24.0844 Ops/s 28.5792 Ops/s $\textbf{\color{#d91a1a}-15.73\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.2780ms 18.0209ms 55.4910 Ops/s 54.4730 Ops/s $\color{#35bf28}+1.87\%$

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens added the Refactoring Refactoring of an existing feature label Nov 6, 2025
vmoens added a commit that referenced this pull request Nov 6, 2025
ghstack-source-id: 83a3415
Pull-Request: #3225
@vmoens vmoens merged commit c02a47c into gh/vmoens/170/base Nov 6, 2025
49 of 61 checks passed
@vmoens vmoens deleted the gh/vmoens/170/head branch November 6, 2025 18:06
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. Refactoring Refactoring of an existing feature

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants