Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 14, 2025

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 14, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3188

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

❌ 5 New Failures, 1 Cancelled Job, 3 Unrelated Failures

As of commit 654200e with merge base 01d2801 (image):

NEW FAILURES - The following jobs have failed:

CANCELLED JOB - The following job was cancelled. Please retry:

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

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

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

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens mentioned this pull request Oct 18, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens added the enhancement New feature or request label Oct 20, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Oct 20, 2025

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}14$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.7783μs 81.4433μs 12.2785 KOps/s 12.2185 KOps/s $\color{#35bf28}+0.49\%$
test_tensor_to_bytestream_speed[torch.save] 0.1416ms 0.1409ms 7.0987 KOps/s 7.0113 KOps/s $\color{#35bf28}+1.25\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1272s 0.1270s 7.8746 Ops/s 7.9665 Ops/s $\color{#d91a1a}-1.15\%$
test_tensor_to_bytestream_speed[numpy] 2.9431μs 2.9364μs 340.5495 KOps/s 358.3561 KOps/s $\color{#d91a1a}-4.97\%$
test_tensor_to_bytestream_speed[safetensors] 42.3199μs 41.6796μs 23.9926 KOps/s 23.7514 KOps/s $\color{#35bf28}+1.02\%$
test_simple 0.5477s 0.5466s 1.8297 Ops/s 1.7351 Ops/s $\textbf{\color{#35bf28}+5.45\%}$
test_transformed 1.2257s 1.1269s 0.8874 Ops/s 0.8828 Ops/s $\color{#35bf28}+0.52\%$
test_serial 1.6594s 1.6551s 0.6042 Ops/s 0.5906 Ops/s $\color{#35bf28}+2.30\%$
test_parallel 1.1752s 1.1103s 0.9007 Ops/s 0.9260 Ops/s $\color{#d91a1a}-2.74\%$
test_step_mdp_speed[True-True-True-True-True] 0.1454ms 43.0648μs 23.2208 KOps/s 22.7363 KOps/s $\color{#35bf28}+2.13\%$
test_step_mdp_speed[True-True-True-True-False] 56.4510μs 24.5326μs 40.7620 KOps/s 39.6676 KOps/s $\color{#35bf28}+2.76\%$
test_step_mdp_speed[True-True-True-False-True] 79.6610μs 24.4797μs 40.8502 KOps/s 40.2054 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[True-True-True-False-False] 39.6000μs 13.6057μs 73.4985 KOps/s 72.2504 KOps/s $\color{#35bf28}+1.73\%$
test_step_mdp_speed[True-True-False-True-True] 78.5310μs 46.9378μs 21.3048 KOps/s 21.0281 KOps/s $\color{#35bf28}+1.32\%$
test_step_mdp_speed[True-True-False-True-False] 89.6710μs 27.6312μs 36.1910 KOps/s 36.1607 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[True-True-False-False-True] 59.9210μs 27.3237μs 36.5982 KOps/s 35.9231 KOps/s $\color{#35bf28}+1.88\%$
test_step_mdp_speed[True-True-False-False-False] 48.9010μs 16.4395μs 60.8290 KOps/s 58.6888 KOps/s $\color{#35bf28}+3.65\%$
test_step_mdp_speed[True-False-True-True-True] 79.8810μs 49.6692μs 20.1332 KOps/s 19.5466 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[True-False-True-True-False] 55.3700μs 30.2199μs 33.0908 KOps/s 32.3928 KOps/s $\color{#35bf28}+2.15\%$
test_step_mdp_speed[True-False-True-False-True] 67.8510μs 27.6594μs 36.1541 KOps/s 36.3065 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[True-False-True-False-False] 48.7610μs 16.3617μs 61.1182 KOps/s 59.5854 KOps/s $\color{#35bf28}+2.57\%$
test_step_mdp_speed[True-False-False-True-True] 96.9910μs 51.9042μs 19.2663 KOps/s 18.8728 KOps/s $\color{#35bf28}+2.09\%$
test_step_mdp_speed[True-False-False-True-False] 61.5010μs 32.6674μs 30.6115 KOps/s 30.3401 KOps/s $\color{#35bf28}+0.89\%$
test_step_mdp_speed[True-False-False-False-True] 57.6010μs 29.7667μs 33.5946 KOps/s 32.6412 KOps/s $\color{#35bf28}+2.92\%$
test_step_mdp_speed[True-False-False-False-False] 51.0410μs 18.9599μs 52.7428 KOps/s 51.9722 KOps/s $\color{#35bf28}+1.48\%$
test_step_mdp_speed[False-True-True-True-True] 79.7510μs 50.2301μs 19.9084 KOps/s 19.8425 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[False-True-True-True-False] 61.7310μs 30.0715μs 33.2541 KOps/s 32.5768 KOps/s $\color{#35bf28}+2.08\%$
test_step_mdp_speed[False-True-True-False-True] 2.6632ms 31.0002μs 32.2579 KOps/s 31.0737 KOps/s $\color{#35bf28}+3.81\%$
test_step_mdp_speed[False-True-True-False-False] 46.3410μs 17.8710μs 55.9564 KOps/s 54.0625 KOps/s $\color{#35bf28}+3.50\%$
test_step_mdp_speed[False-True-False-True-True] 90.4110μs 51.8873μs 19.2726 KOps/s 18.7833 KOps/s $\color{#35bf28}+2.60\%$
test_step_mdp_speed[False-True-False-True-False] 64.7310μs 32.3583μs 30.9039 KOps/s 30.3646 KOps/s $\color{#35bf28}+1.78\%$
test_step_mdp_speed[False-True-False-False-True] 62.6910μs 33.1543μs 30.1620 KOps/s 29.4658 KOps/s $\color{#35bf28}+2.36\%$
test_step_mdp_speed[False-True-False-False-False] 64.1010μs 20.6067μs 48.5280 KOps/s 47.7952 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-False-True-True-True] 93.4020μs 54.9450μs 18.2000 KOps/s 17.9422 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[False-False-True-True-False] 0.1364ms 34.6882μs 28.8283 KOps/s 27.8905 KOps/s $\color{#35bf28}+3.36\%$
test_step_mdp_speed[False-False-True-False-True] 66.4010μs 33.4719μs 29.8758 KOps/s 29.1863 KOps/s $\color{#35bf28}+2.36\%$
test_step_mdp_speed[False-False-True-False-False] 48.9000μs 20.5267μs 48.7170 KOps/s 48.1443 KOps/s $\color{#35bf28}+1.19\%$
test_step_mdp_speed[False-False-False-True-True] 86.5220μs 56.0298μs 17.8477 KOps/s 17.4609 KOps/s $\color{#35bf28}+2.22\%$
test_step_mdp_speed[False-False-False-True-False] 65.8310μs 37.9925μs 26.3210 KOps/s 26.0800 KOps/s $\color{#35bf28}+0.92\%$
test_step_mdp_speed[False-False-False-False-True] 58.0010μs 35.5266μs 28.1479 KOps/s 27.5852 KOps/s $\color{#35bf28}+2.04\%$
test_step_mdp_speed[False-False-False-False-False] 70.4110μs 23.0155μs 43.4489 KOps/s 43.1842 KOps/s $\color{#35bf28}+0.61\%$
test_values[generalized_advantage_estimate-True-True] 10.4377ms 10.0986ms 99.0240 Ops/s 99.8359 Ops/s $\color{#d91a1a}-0.81\%$
test_values[vec_generalized_advantage_estimate-True-True] 16.4606ms 11.3388ms 88.1928 Ops/s 55.6419 Ops/s $\textbf{\color{#35bf28}+58.50\%}$
test_values[td0_return_estimate-False-False] 0.2427ms 0.1336ms 7.4840 KOps/s 7.8100 KOps/s $\color{#d91a1a}-4.17\%$
test_values[td1_return_estimate-False-False] 27.6687ms 27.1723ms 36.8022 Ops/s 37.3446 Ops/s $\color{#d91a1a}-1.45\%$
test_values[vec_td1_return_estimate-False-False] 12.1178ms 11.3714ms 87.9399 Ops/s 56.6090 Ops/s $\textbf{\color{#35bf28}+55.35\%}$
test_values[td_lambda_return_estimate-True-False] 41.4571ms 40.6886ms 24.5769 Ops/s 24.7453 Ops/s $\color{#d91a1a}-0.68\%$
test_values[vec_td_lambda_return_estimate-True-False] 12.5746ms 11.5001ms 86.9560 Ops/s 56.7270 Ops/s $\textbf{\color{#35bf28}+53.29\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.8597ms 8.6478ms 115.6367 Ops/s 116.4468 Ops/s $\color{#d91a1a}-0.70\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8929ms 1.5139ms 660.5261 Ops/s 643.1033 Ops/s $\color{#35bf28}+2.71\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4830ms 0.4176ms 2.3944 KOps/s 2.3941 KOps/s $+0.01\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 30.5664ms 28.9748ms 34.5128 Ops/s 28.7454 Ops/s $\textbf{\color{#35bf28}+20.06\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.1235ms 1.7313ms 577.5862 Ops/s 574.0877 Ops/s $\color{#35bf28}+0.61\%$
test_dqn_speed[False-None] 1.5183ms 1.4235ms 702.5153 Ops/s 708.8555 Ops/s $\color{#d91a1a}-0.89\%$
test_dqn_speed[False-backward] 2.0174ms 1.9342ms 516.9975 Ops/s 517.6991 Ops/s $\color{#d91a1a}-0.14\%$
test_dqn_speed[True-None] 1.1943ms 0.5391ms 1.8548 KOps/s 1.8871 KOps/s $\color{#d91a1a}-1.71\%$
test_dqn_speed[True-backward] 1.0321ms 0.9827ms 1.0176 KOps/s 996.4293 Ops/s $\color{#35bf28}+2.12\%$
test_dqn_speed[reduce-overhead-None] 0.9302ms 0.5209ms 1.9198 KOps/s 1.9062 KOps/s $\color{#35bf28}+0.71\%$
test_dqn_speed[reduce-overhead-backward] 1.0661ms 0.9767ms 1.0239 KOps/s 1.0169 KOps/s $\color{#35bf28}+0.69\%$
test_ddpg_speed[False-None] 3.2954ms 2.9197ms 342.5027 Ops/s 344.9394 Ops/s $\color{#d91a1a}-0.71\%$
test_ddpg_speed[False-backward] 4.9773ms 4.1482ms 241.0703 Ops/s 241.4894 Ops/s $\color{#d91a1a}-0.17\%$
test_ddpg_speed[True-None] 5.9005ms 1.4026ms 712.9570 Ops/s 664.3493 Ops/s $\textbf{\color{#35bf28}+7.32\%}$
test_ddpg_speed[True-backward] 2.5506ms 2.4363ms 410.4646 Ops/s 349.9893 Ops/s $\textbf{\color{#35bf28}+17.28\%}$
test_ddpg_speed[reduce-overhead-None] 1.5195ms 1.3920ms 718.3986 Ops/s 720.4898 Ops/s $\color{#d91a1a}-0.29\%$
test_ddpg_speed[reduce-overhead-backward] 2.4941ms 2.3798ms 420.2073 Ops/s 340.0793 Ops/s $\textbf{\color{#35bf28}+23.56\%}$
test_sac_speed[False-None] 8.5243ms 7.9412ms 125.9250 Ops/s 123.7719 Ops/s $\color{#35bf28}+1.74\%$
test_sac_speed[False-backward] 11.9156ms 11.4210ms 87.5579 Ops/s 89.2091 Ops/s $\color{#d91a1a}-1.85\%$
test_sac_speed[True-None] 2.5166ms 2.1696ms 460.9176 Ops/s 459.1152 Ops/s $\color{#35bf28}+0.39\%$
test_sac_speed[True-backward] 4.2926ms 4.1382ms 241.6507 Ops/s 239.9685 Ops/s $\color{#35bf28}+0.70\%$
test_sac_speed[reduce-overhead-None] 2.7149ms 2.1756ms 459.6536 Ops/s 450.3840 Ops/s $\color{#35bf28}+2.06\%$
test_sac_speed[reduce-overhead-backward] 4.3818ms 4.1710ms 239.7478 Ops/s 218.8814 Ops/s $\textbf{\color{#35bf28}+9.53\%}$
test_redq_speed[False-None] 11.1424ms 10.5839ms 94.4832 Ops/s 91.6235 Ops/s $\color{#35bf28}+3.12\%$
test_redq_speed[False-backward] 19.2648ms 18.6065ms 53.7446 Ops/s 54.2612 Ops/s $\color{#d91a1a}-0.95\%$
test_redq_speed[True-None] 5.0842ms 4.7267ms 211.5641 Ops/s 213.2328 Ops/s $\color{#d91a1a}-0.78\%$
test_redq_speed[True-backward] 12.2059ms 10.2271ms 97.7790 Ops/s 99.8467 Ops/s $\color{#d91a1a}-2.07\%$
test_redq_speed[reduce-overhead-None] 5.0282ms 4.5471ms 219.9189 Ops/s 209.8366 Ops/s $\color{#35bf28}+4.80\%$
test_redq_speed[reduce-overhead-backward] 10.7727ms 10.3920ms 96.2280 Ops/s 99.0903 Ops/s $\color{#d91a1a}-2.89\%$
test_redq_deprec_speed[False-None] 11.5985ms 11.2244ms 89.0917 Ops/s 90.4176 Ops/s $\color{#d91a1a}-1.47\%$
test_redq_deprec_speed[False-backward] 17.0264ms 16.3184ms 61.2805 Ops/s 62.4528 Ops/s $\color{#d91a1a}-1.88\%$
test_redq_deprec_speed[True-None] 4.1054ms 3.7493ms 266.7180 Ops/s 256.9978 Ops/s $\color{#35bf28}+3.78\%$
test_redq_deprec_speed[True-backward] 8.4299ms 7.8890ms 126.7586 Ops/s 121.7952 Ops/s $\color{#35bf28}+4.08\%$
test_redq_deprec_speed[reduce-overhead-None] 4.0929ms 3.6897ms 271.0282 Ops/s 265.2000 Ops/s $\color{#35bf28}+2.20\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.1948ms 7.9780ms 125.3454 Ops/s 124.4971 Ops/s $\color{#35bf28}+0.68\%$
test_td3_speed[False-None] 8.1636ms 8.0454ms 124.2946 Ops/s 119.1778 Ops/s $\color{#35bf28}+4.29\%$
test_td3_speed[False-backward] 11.7151ms 11.1145ms 89.9722 Ops/s 91.9514 Ops/s $\color{#d91a1a}-2.15\%$
test_td3_speed[True-None] 1.9741ms 1.8883ms 529.5693 Ops/s 534.6483 Ops/s $\color{#d91a1a}-0.95\%$
test_td3_speed[True-backward] 3.8503ms 3.7211ms 268.7388 Ops/s 241.1937 Ops/s $\textbf{\color{#35bf28}+11.42\%}$
test_td3_speed[reduce-overhead-None] 1.9078ms 1.8573ms 538.4139 Ops/s 542.3635 Ops/s $\color{#d91a1a}-0.73\%$
test_td3_speed[reduce-overhead-backward] 4.1036ms 3.8047ms 262.8334 Ops/s 257.4123 Ops/s $\color{#35bf28}+2.11\%$
test_cql_speed[False-None] 29.4005ms 26.7850ms 37.3343 Ops/s 37.2650 Ops/s $\color{#35bf28}+0.19\%$
test_cql_speed[False-backward] 38.5257ms 36.3489ms 27.5112 Ops/s 27.2946 Ops/s $\color{#35bf28}+0.79\%$
test_cql_speed[True-None] 13.3907ms 12.7753ms 78.2759 Ops/s 75.9565 Ops/s $\color{#35bf28}+3.05\%$
test_cql_speed[True-backward] 19.5788ms 18.9521ms 52.7646 Ops/s 54.0722 Ops/s $\color{#d91a1a}-2.42\%$
test_cql_speed[reduce-overhead-None] 13.2303ms 12.8104ms 78.0616 Ops/s 78.4607 Ops/s $\color{#d91a1a}-0.51\%$
test_cql_speed[reduce-overhead-backward] 19.6486ms 19.0000ms 52.6316 Ops/s 54.4084 Ops/s $\color{#d91a1a}-3.27\%$
test_a2c_speed[False-None] 6.0881ms 5.5487ms 180.2221 Ops/s 180.1731 Ops/s $\color{#35bf28}+0.03\%$
test_a2c_speed[False-backward] 12.7470ms 12.1847ms 82.0700 Ops/s 83.2385 Ops/s $\color{#d91a1a}-1.40\%$
test_a2c_speed[True-None] 4.2901ms 3.8031ms 262.9412 Ops/s 260.9535 Ops/s $\color{#35bf28}+0.76\%$
test_a2c_speed[True-backward] 9.1042ms 8.8441ms 113.0693 Ops/s 105.8512 Ops/s $\textbf{\color{#35bf28}+6.82\%}$
test_a2c_speed[reduce-overhead-None] 4.3006ms 3.8374ms 260.5908 Ops/s 267.0480 Ops/s $\color{#d91a1a}-2.42\%$
test_a2c_speed[reduce-overhead-backward] 9.4048ms 9.1379ms 109.4341 Ops/s 108.1724 Ops/s $\color{#35bf28}+1.17\%$
test_ppo_speed[False-None] 6.4474ms 6.0612ms 164.9835 Ops/s 164.0276 Ops/s $\color{#35bf28}+0.58\%$
test_ppo_speed[False-backward] 13.6858ms 13.0841ms 76.4286 Ops/s 76.9211 Ops/s $\color{#d91a1a}-0.64\%$
test_ppo_speed[True-None] 3.9191ms 3.7272ms 268.2952 Ops/s 269.8237 Ops/s $\color{#d91a1a}-0.57\%$
test_ppo_speed[True-backward] 9.2121ms 8.7656ms 114.0829 Ops/s 115.1713 Ops/s $\color{#d91a1a}-0.95\%$
test_ppo_speed[reduce-overhead-None] 3.8984ms 3.7433ms 267.1415 Ops/s 272.0186 Ops/s $\color{#d91a1a}-1.79\%$
test_ppo_speed[reduce-overhead-backward] 9.3986ms 8.9940ms 111.1847 Ops/s 111.0232 Ops/s $\color{#35bf28}+0.15\%$
test_reinforce_speed[False-None] 5.0654ms 4.7232ms 211.7203 Ops/s 212.4045 Ops/s $\color{#d91a1a}-0.32\%$
test_reinforce_speed[False-backward] 7.8843ms 7.6451ms 130.8019 Ops/s 131.7829 Ops/s $\color{#d91a1a}-0.74\%$
test_reinforce_speed[True-None] 3.3925ms 2.9783ms 335.7613 Ops/s 339.8325 Ops/s $\color{#d91a1a}-1.20\%$
test_reinforce_speed[True-backward] 8.4696ms 7.9478ms 125.8215 Ops/s 128.2364 Ops/s $\color{#d91a1a}-1.88\%$
test_reinforce_speed[reduce-overhead-None] 3.5044ms 2.9511ms 338.8603 Ops/s 328.2882 Ops/s $\color{#35bf28}+3.22\%$
test_reinforce_speed[reduce-overhead-backward] 8.5457ms 8.1731ms 122.3527 Ops/s 120.1855 Ops/s $\color{#35bf28}+1.80\%$
test_iql_speed[False-None] 25.5006ms 20.9281ms 47.7826 Ops/s 47.3262 Ops/s $\color{#35bf28}+0.96\%$
test_iql_speed[False-backward] 37.0261ms 31.7070ms 31.5387 Ops/s 31.3913 Ops/s $\color{#35bf28}+0.47\%$
test_iql_speed[True-None] 9.1901ms 8.7702ms 114.0221 Ops/s 112.6333 Ops/s $\color{#35bf28}+1.23\%$
test_iql_speed[True-backward] 17.8956ms 17.2840ms 57.8571 Ops/s 57.7736 Ops/s $\color{#35bf28}+0.14\%$
test_iql_speed[reduce-overhead-None] 9.2250ms 8.8576ms 112.8971 Ops/s 101.1576 Ops/s $\textbf{\color{#35bf28}+11.61\%}$
test_iql_speed[reduce-overhead-backward] 18.2394ms 17.7791ms 56.2457 Ops/s 56.8737 Ops/s $\color{#d91a1a}-1.10\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5082ms 6.0508ms 165.2660 Ops/s 167.8006 Ops/s $\color{#d91a1a}-1.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5903ms 0.3129ms 3.1958 KOps/s 3.4974 KOps/s $\textbf{\color{#d91a1a}-8.62\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7019ms 0.2966ms 3.3710 KOps/s 3.7862 KOps/s $\textbf{\color{#d91a1a}-10.96\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9525ms 5.7244ms 174.6920 Ops/s 176.3880 Ops/s $\color{#d91a1a}-0.96\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0528ms 0.3269ms 3.0588 KOps/s 3.3206 KOps/s $\textbf{\color{#d91a1a}-7.88\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6690ms 0.3251ms 3.0755 KOps/s 3.8050 KOps/s $\textbf{\color{#d91a1a}-19.17\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6936ms 1.4522ms 688.6236 Ops/s 790.4668 Ops/s $\textbf{\color{#d91a1a}-12.88\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.7932ms 1.3623ms 734.0330 Ops/s 841.8463 Ops/s $\textbf{\color{#d91a1a}-12.81\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.9419ms 6.1842ms 161.7017 Ops/s 170.9303 Ops/s $\textbf{\color{#d91a1a}-5.40\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9335ms 0.5329ms 1.8765 KOps/s 1.9972 KOps/s $\textbf{\color{#d91a1a}-6.04\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7652ms 0.5143ms 1.9442 KOps/s 2.0665 KOps/s $\textbf{\color{#d91a1a}-5.92\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9139ms 5.7395ms 174.2299 Ops/s 176.8036 Ops/s $\color{#d91a1a}-1.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0955ms 0.3495ms 2.8609 KOps/s 765.9590 Ops/s $\textbf{\color{#35bf28}+273.50\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5558ms 0.3633ms 2.7523 KOps/s 3.7668 KOps/s $\textbf{\color{#d91a1a}-26.93\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0466ms 5.7170ms 174.9154 Ops/s 174.4321 Ops/s $\color{#35bf28}+0.28\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6775ms 0.3437ms 2.9099 KOps/s 3.5673 KOps/s $\textbf{\color{#d91a1a}-18.43\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5779ms 0.3181ms 3.1436 KOps/s 3.8307 KOps/s $\textbf{\color{#d91a1a}-17.94\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1204ms 5.9120ms 169.1488 Ops/s 167.9258 Ops/s $\color{#35bf28}+0.73\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0209ms 0.4803ms 2.0822 KOps/s 2.2688 KOps/s $\textbf{\color{#d91a1a}-8.22\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7043ms 0.4630ms 2.1598 KOps/s 2.4071 KOps/s $\textbf{\color{#d91a1a}-10.27\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.5568ms 5.0397ms 198.4264 Ops/s 194.8688 Ops/s $\color{#35bf28}+1.83\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.1816ms 2.3235ms 430.3776 Ops/s 430.5643 Ops/s $\color{#d91a1a}-0.04\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.1891ms 1.2634ms 791.5359 Ops/s 806.9852 Ops/s $\color{#d91a1a}-1.91\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5136s 15.2635ms 65.5158 Ops/s 56.2376 Ops/s $\textbf{\color{#35bf28}+16.50\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 11.6708ms 1.9478ms 513.3969 Ops/s 482.0358 Ops/s $\textbf{\color{#35bf28}+6.51\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.4469ms 1.1861ms 843.1119 Ops/s 852.0853 Ops/s $\color{#d91a1a}-1.05\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.7282ms 5.2411ms 190.7989 Ops/s 189.0141 Ops/s $\color{#35bf28}+0.94\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.8600ms 2.2423ms 445.9627 Ops/s 466.1618 Ops/s $\color{#d91a1a}-4.33\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 3.7245ms 1.2582ms 794.8147 Ops/s 710.4110 Ops/s $\textbf{\color{#35bf28}+11.88\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 34.8094ms 32.6412ms 30.6361 Ops/s 30.3201 Ops/s $\color{#35bf28}+1.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.4515ms 17.8073ms 56.1568 Ops/s 57.4918 Ops/s $\color{#d91a1a}-2.32\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 36.7578ms 34.1710ms 29.2646 Ops/s 29.4051 Ops/s $\color{#d91a1a}-0.48\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.0116ms 17.6463ms 56.6692 Ops/s 56.2139 Ops/s $\color{#35bf28}+0.81\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 37.8901ms 36.1686ms 27.6483 Ops/s 28.1393 Ops/s $\color{#d91a1a}-1.74\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.5262ms 19.3098ms 51.7872 Ops/s 52.5179 Ops/s $\color{#d91a1a}-1.39\%$

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added 11 commits October 22, 2025 12:31
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 25, 2025
@vmoens vmoens merged commit 654200e into gh/vmoens/151/base Oct 25, 2025
92 of 101 checks passed
@vmoens vmoens deleted the gh/vmoens/151/head branch October 25, 2025 00:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. enhancement New feature or request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant