Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 18, 2025

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Oct 18, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3211

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

⏳ No Failures, 32 Pending

As of commit 2e9e545 with merge base 92c20cd (image):
💚 Looks good so far! There are no failures yet. 💚

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

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 18, 2025
ghstack-source-id: e41e8b1
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 19, 2025
ghstack-source-id: f9b82b5
Pull-Request: #3211
vmoens added a commit that referenced this pull request Oct 19, 2025
ghstack-source-id: f9b82b5
Pull-Request: #3211
vmoens added a commit that referenced this pull request Oct 19, 2025
ghstack-source-id: f9b82b5
Pull-Request: #3211
vmoens added a commit that referenced this pull request Oct 19, 2025
ghstack-source-id: f9b82b5
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: fa649fd
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: ab51147
Pull-Request: #3211
@vmoens vmoens added the quality code quality label Oct 20, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: 609aeb2
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: b56e446
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: 2b57e12
Pull-Request: #3211
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: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}23$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.3429μs 80.0240μs 12.4963 KOps/s 12.1328 KOps/s $\color{#35bf28}+3.00\%$
test_tensor_to_bytestream_speed[torch.save] 0.1365ms 0.1357ms 7.3702 KOps/s 7.2696 KOps/s $\color{#35bf28}+1.38\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1126s 0.1123s 8.9044 Ops/s 9.1607 Ops/s $\color{#d91a1a}-2.80\%$
test_tensor_to_bytestream_speed[numpy] 2.7311μs 2.7278μs 366.5999 KOps/s 370.2432 KOps/s $\color{#d91a1a}-0.98\%$
test_tensor_to_bytestream_speed[safetensors] 42.0903μs 41.9070μs 23.8624 KOps/s 23.3329 KOps/s $\color{#35bf28}+2.27\%$
test_simple 0.5326s 0.5306s 1.8847 Ops/s 1.7935 Ops/s $\textbf{\color{#35bf28}+5.08\%}$
test_transformed 1.0792s 1.0769s 0.9286 Ops/s 0.9032 Ops/s $\color{#35bf28}+2.81\%$
test_serial 1.6040s 1.6006s 0.6248 Ops/s 0.6139 Ops/s $\color{#35bf28}+1.77\%$
test_parallel 1.0499s 1.0375s 0.9639 Ops/s 0.9511 Ops/s $\color{#35bf28}+1.34\%$
test_step_mdp_speed[True-True-True-True-True] 0.1628ms 42.3161μs 23.6317 KOps/s 22.5026 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_step_mdp_speed[True-True-True-True-False] 49.2010μs 24.3700μs 41.0340 KOps/s 39.8827 KOps/s $\color{#35bf28}+2.89\%$
test_step_mdp_speed[True-True-True-False-True] 53.0700μs 24.4016μs 40.9810 KOps/s 40.4667 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[True-True-True-False-False] 38.8010μs 13.4833μs 74.1657 KOps/s 72.9797 KOps/s $\color{#35bf28}+1.63\%$
test_step_mdp_speed[True-True-False-True-True] 70.9620μs 46.2900μs 21.6029 KOps/s 21.5986 KOps/s $\color{#35bf28}+0.02\%$
test_step_mdp_speed[True-True-False-True-False] 57.6910μs 26.5420μs 37.6762 KOps/s 36.4182 KOps/s $\color{#35bf28}+3.45\%$
test_step_mdp_speed[True-True-False-False-True] 55.6210μs 26.5217μs 37.7049 KOps/s 37.3298 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[True-True-False-False-False] 51.0410μs 16.0359μs 62.3603 KOps/s 62.2738 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[True-False-True-True-True] 82.5010μs 47.4381μs 21.0801 KOps/s 20.0805 KOps/s $\color{#35bf28}+4.98\%$
test_step_mdp_speed[True-False-True-True-False] 62.7110μs 29.1767μs 34.2740 KOps/s 33.3211 KOps/s $\color{#35bf28}+2.86\%$
test_step_mdp_speed[True-False-True-False-True] 63.4410μs 26.9120μs 37.1582 KOps/s 36.7463 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[True-False-True-False-False] 45.3300μs 15.8577μs 63.0610 KOps/s 62.0875 KOps/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[True-False-False-True-True] 84.0810μs 50.2708μs 19.8923 KOps/s 19.2216 KOps/s $\color{#35bf28}+3.49\%$
test_step_mdp_speed[True-False-False-True-False] 65.8910μs 31.8841μs 31.3636 KOps/s 31.4539 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[True-False-False-False-True] 54.0410μs 29.6149μs 33.7668 KOps/s 33.9100 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[True-False-False-False-False] 75.5210μs 18.7396μs 53.3629 KOps/s 53.1391 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[False-True-True-True-True] 81.9220μs 48.5251μs 20.6079 KOps/s 20.3156 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[False-True-True-True-False] 56.7510μs 29.6549μs 33.7212 KOps/s 33.6330 KOps/s $\color{#35bf28}+0.26\%$
test_step_mdp_speed[False-True-True-False-True] 2.4329ms 30.9191μs 32.3424 KOps/s 31.6490 KOps/s $\color{#35bf28}+2.19\%$
test_step_mdp_speed[False-True-True-False-False] 74.1810μs 17.0642μs 58.6020 KOps/s 54.7692 KOps/s $\textbf{\color{#35bf28}+7.00\%}$
test_step_mdp_speed[False-True-False-True-True] 93.4220μs 51.6477μs 19.3619 KOps/s 19.1816 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[False-True-False-True-False] 65.4210μs 32.2615μs 30.9967 KOps/s 30.1408 KOps/s $\color{#35bf28}+2.84\%$
test_step_mdp_speed[False-True-False-False-True] 70.1310μs 32.6011μs 30.6738 KOps/s 29.1821 KOps/s $\textbf{\color{#35bf28}+5.11\%}$
test_step_mdp_speed[False-True-False-False-False] 56.4510μs 20.2753μs 49.3211 KOps/s 46.8909 KOps/s $\textbf{\color{#35bf28}+5.18\%}$
test_step_mdp_speed[False-False-True-True-True] 88.6610μs 53.8881μs 18.5570 KOps/s 18.0160 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[False-False-True-True-False] 67.7120μs 34.3651μs 29.0993 KOps/s 27.9089 KOps/s $\color{#35bf28}+4.27\%$
test_step_mdp_speed[False-False-True-False-True] 67.8710μs 32.6430μs 30.6344 KOps/s 28.9190 KOps/s $\textbf{\color{#35bf28}+5.93\%}$
test_step_mdp_speed[False-False-True-False-False] 57.7810μs 20.6098μs 48.5205 KOps/s 48.0335 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[False-False-False-True-True] 87.5220μs 55.8847μs 17.8940 KOps/s 17.5452 KOps/s $\color{#35bf28}+1.99\%$
test_step_mdp_speed[False-False-False-True-False] 78.7420μs 36.9766μs 27.0441 KOps/s 26.2182 KOps/s $\color{#35bf28}+3.15\%$
test_step_mdp_speed[False-False-False-False-True] 76.5110μs 34.8980μs 28.6550 KOps/s 28.0698 KOps/s $\color{#35bf28}+2.08\%$
test_step_mdp_speed[False-False-False-False-False] 59.3710μs 22.5726μs 44.3015 KOps/s 43.2893 KOps/s $\color{#35bf28}+2.34\%$
test_values[generalized_advantage_estimate-True-True] 10.1445ms 9.4068ms 106.3064 Ops/s 104.3719 Ops/s $\color{#35bf28}+1.85\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.7497ms 17.5592ms 56.9502 Ops/s 90.3946 Ops/s $\textbf{\color{#d91a1a}-37.00\%}$
test_values[td0_return_estimate-False-False] 0.2176ms 0.1287ms 7.7715 KOps/s 7.9065 KOps/s $\color{#d91a1a}-1.71\%$
test_values[td1_return_estimate-False-False] 27.6491ms 26.3495ms 37.9514 Ops/s 38.6565 Ops/s $\color{#d91a1a}-1.82\%$
test_values[vec_td1_return_estimate-False-False] 18.6686ms 17.6783ms 56.5666 Ops/s 87.4944 Ops/s $\textbf{\color{#d91a1a}-35.35\%}$
test_values[td_lambda_return_estimate-True-False] 41.7216ms 40.1014ms 24.9368 Ops/s 25.7980 Ops/s $\color{#d91a1a}-3.34\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.2375ms 17.7071ms 56.4745 Ops/s 90.1625 Ops/s $\textbf{\color{#d91a1a}-37.36\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.1627ms 8.0947ms 123.5377 Ops/s 125.4996 Ops/s $\color{#d91a1a}-1.56\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8039ms 1.5229ms 656.6632 Ops/s 667.5776 Ops/s $\color{#d91a1a}-1.63\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4713ms 0.3986ms 2.5086 KOps/s 2.5173 KOps/s $\color{#d91a1a}-0.34\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.8990ms 33.9833ms 29.4262 Ops/s 33.7450 Ops/s $\textbf{\color{#d91a1a}-12.80\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9620ms 1.7036ms 586.9885 Ops/s 583.1734 Ops/s $\color{#35bf28}+0.65\%$
test_dqn_speed[False-None] 6.4645ms 1.3821ms 723.5400 Ops/s 713.1479 Ops/s $\color{#35bf28}+1.46\%$
test_dqn_speed[False-backward] 2.0210ms 1.8822ms 531.2870 Ops/s 515.0397 Ops/s $\color{#35bf28}+3.15\%$
test_dqn_speed[True-None] 0.7186ms 0.5292ms 1.8898 KOps/s 1.9572 KOps/s $\color{#d91a1a}-3.45\%$
test_dqn_speed[True-backward] 0.9999ms 0.9501ms 1.0525 KOps/s 1.0145 KOps/s $\color{#35bf28}+3.74\%$
test_dqn_speed[reduce-overhead-None] 0.9117ms 0.5154ms 1.9402 KOps/s 1.9782 KOps/s $\color{#d91a1a}-1.92\%$
test_dqn_speed[reduce-overhead-backward] 1.0100ms 0.9375ms 1.0666 KOps/s 1.0419 KOps/s $\color{#35bf28}+2.37\%$
test_ddpg_speed[False-None] 3.1672ms 2.7776ms 360.0204 Ops/s 358.1877 Ops/s $\color{#35bf28}+0.51\%$
test_ddpg_speed[False-backward] 4.1497ms 4.0295ms 248.1676 Ops/s 249.9197 Ops/s $\color{#d91a1a}-0.70\%$
test_ddpg_speed[True-None] 1.7294ms 1.3920ms 718.3850 Ops/s 727.7992 Ops/s $\color{#d91a1a}-1.29\%$
test_ddpg_speed[True-backward] 2.3824ms 2.3344ms 428.3693 Ops/s 398.4147 Ops/s $\textbf{\color{#35bf28}+7.52\%}$
test_ddpg_speed[reduce-overhead-None] 1.7941ms 1.3842ms 722.4552 Ops/s 712.5490 Ops/s $\color{#35bf28}+1.39\%$
test_ddpg_speed[reduce-overhead-backward] 2.4782ms 2.3189ms 431.2363 Ops/s 430.8265 Ops/s $\color{#35bf28}+0.10\%$
test_sac_speed[False-None] 8.1610ms 7.6813ms 130.1868 Ops/s 129.4056 Ops/s $\color{#35bf28}+0.60\%$
test_sac_speed[False-backward] 11.3561ms 10.9163ms 91.6058 Ops/s 91.1441 Ops/s $\color{#35bf28}+0.51\%$
test_sac_speed[True-None] 2.4386ms 2.0742ms 482.1212 Ops/s 464.4558 Ops/s $\color{#35bf28}+3.80\%$
test_sac_speed[True-backward] 4.0473ms 3.9334ms 254.2313 Ops/s 250.0404 Ops/s $\color{#35bf28}+1.68\%$
test_sac_speed[reduce-overhead-None] 2.4483ms 2.0650ms 484.2683 Ops/s 452.5900 Ops/s $\textbf{\color{#35bf28}+7.00\%}$
test_sac_speed[reduce-overhead-backward] 4.1070ms 3.9364ms 254.0391 Ops/s 222.9681 Ops/s $\textbf{\color{#35bf28}+13.94\%}$
test_redq_speed[False-None] 15.5332ms 10.2068ms 97.9743 Ops/s 99.2583 Ops/s $\color{#d91a1a}-1.29\%$
test_redq_speed[False-backward] 17.8736ms 17.2016ms 58.1343 Ops/s 57.8255 Ops/s $\color{#35bf28}+0.53\%$
test_redq_speed[True-None] 4.5412ms 4.3277ms 231.0676 Ops/s 226.8624 Ops/s $\color{#35bf28}+1.85\%$
test_redq_speed[True-backward] 9.8681ms 9.5018ms 105.2431 Ops/s 103.7840 Ops/s $\color{#35bf28}+1.41\%$
test_redq_speed[reduce-overhead-None] 5.2342ms 4.3404ms 230.3956 Ops/s 234.8153 Ops/s $\color{#d91a1a}-1.88\%$
test_redq_speed[reduce-overhead-backward] 10.1546ms 9.7673ms 102.3822 Ops/s 105.2226 Ops/s $\color{#d91a1a}-2.70\%$
test_redq_deprec_speed[False-None] 11.0779ms 10.6378ms 94.0041 Ops/s 90.8470 Ops/s $\color{#35bf28}+3.48\%$
test_redq_deprec_speed[False-backward] 16.7503ms 15.4033ms 64.9213 Ops/s 62.4177 Ops/s $\color{#35bf28}+4.01\%$
test_redq_deprec_speed[True-None] 3.6663ms 3.5289ms 283.3782 Ops/s 266.5274 Ops/s $\textbf{\color{#35bf28}+6.32\%}$
test_redq_deprec_speed[True-backward] 7.5310ms 7.2868ms 137.2347 Ops/s 129.4623 Ops/s $\textbf{\color{#35bf28}+6.00\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.6932ms 3.5092ms 284.9681 Ops/s 290.2853 Ops/s $\color{#d91a1a}-1.83\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.5452ms 7.2962ms 137.0584 Ops/s 130.2627 Ops/s $\textbf{\color{#35bf28}+5.22\%}$
test_td3_speed[False-None] 7.9572ms 7.7255ms 129.4415 Ops/s 130.2884 Ops/s $\color{#d91a1a}-0.65\%$
test_td3_speed[False-backward] 11.2774ms 10.5735ms 94.5759 Ops/s 94.9652 Ops/s $\color{#d91a1a}-0.41\%$
test_td3_speed[True-None] 1.8517ms 1.7822ms 561.1182 Ops/s 558.4422 Ops/s $\color{#35bf28}+0.48\%$
test_td3_speed[True-backward] 3.7284ms 3.5561ms 281.2089 Ops/s 274.8266 Ops/s $\color{#35bf28}+2.32\%$
test_td3_speed[reduce-overhead-None] 1.7973ms 1.7555ms 569.6395 Ops/s 560.4733 Ops/s $\color{#35bf28}+1.64\%$
test_td3_speed[reduce-overhead-backward] 3.7320ms 3.5687ms 280.2110 Ops/s 237.0913 Ops/s $\textbf{\color{#35bf28}+18.19\%}$
test_cql_speed[False-None] 25.9096ms 25.1953ms 39.6899 Ops/s 39.5279 Ops/s $\color{#35bf28}+0.41\%$
test_cql_speed[False-backward] 39.8606ms 35.1960ms 28.4123 Ops/s 28.6943 Ops/s $\color{#d91a1a}-0.98\%$
test_cql_speed[True-None] 12.5403ms 12.2481ms 81.6452 Ops/s 85.2125 Ops/s $\color{#d91a1a}-4.19\%$
test_cql_speed[True-backward] 18.3278ms 17.9768ms 55.6272 Ops/s 59.2643 Ops/s $\textbf{\color{#d91a1a}-6.14\%}$
test_cql_speed[reduce-overhead-None] 12.6634ms 12.3288ms 81.1110 Ops/s 84.8083 Ops/s $\color{#d91a1a}-4.36\%$
test_cql_speed[reduce-overhead-backward] 18.4733ms 18.0443ms 55.4191 Ops/s 53.4888 Ops/s $\color{#35bf28}+3.61\%$
test_a2c_speed[False-None] 5.6698ms 5.2858ms 189.1867 Ops/s 186.3218 Ops/s $\color{#35bf28}+1.54\%$
test_a2c_speed[False-backward] 12.0280ms 11.7517ms 85.0944 Ops/s 84.6739 Ops/s $\color{#35bf28}+0.50\%$
test_a2c_speed[True-None] 3.9258ms 3.6562ms 273.5075 Ops/s 280.0569 Ops/s $\color{#d91a1a}-2.34\%$
test_a2c_speed[True-backward] 9.8616ms 8.6360ms 115.7939 Ops/s 121.5099 Ops/s $\color{#d91a1a}-4.70\%$
test_a2c_speed[reduce-overhead-None] 4.1250ms 3.6881ms 271.1396 Ops/s 289.9224 Ops/s $\textbf{\color{#d91a1a}-6.48\%}$
test_a2c_speed[reduce-overhead-backward] 9.0236ms 8.8191ms 113.3903 Ops/s 118.6969 Ops/s $\color{#d91a1a}-4.47\%$
test_ppo_speed[False-None] 6.0433ms 5.7383ms 174.2681 Ops/s 179.9105 Ops/s $\color{#d91a1a}-3.14\%$
test_ppo_speed[False-backward] 12.7873ms 12.2044ms 81.9377 Ops/s 81.7590 Ops/s $\color{#35bf28}+0.22\%$
test_ppo_speed[True-None] 3.8202ms 3.6451ms 274.3375 Ops/s 289.7643 Ops/s $\textbf{\color{#d91a1a}-5.32\%}$
test_ppo_speed[True-backward] 8.6001ms 8.3300ms 120.0476 Ops/s 115.3532 Ops/s $\color{#35bf28}+4.07\%$
test_ppo_speed[reduce-overhead-None] 3.8216ms 3.6278ms 275.6473 Ops/s 295.2974 Ops/s $\textbf{\color{#d91a1a}-6.65\%}$
test_ppo_speed[reduce-overhead-backward] 8.7967ms 8.6065ms 116.1911 Ops/s 119.3717 Ops/s $\color{#d91a1a}-2.66\%$
test_reinforce_speed[False-None] 4.8882ms 4.4874ms 222.8479 Ops/s 237.4915 Ops/s $\textbf{\color{#d91a1a}-6.17\%}$
test_reinforce_speed[False-backward] 7.7454ms 7.2960ms 137.0609 Ops/s 141.9642 Ops/s $\color{#d91a1a}-3.45\%$
test_reinforce_speed[True-None] 2.9882ms 2.8293ms 353.4445 Ops/s 383.0656 Ops/s $\textbf{\color{#d91a1a}-7.73\%}$
test_reinforce_speed[True-backward] 7.8970ms 7.5714ms 132.0755 Ops/s 128.1198 Ops/s $\color{#35bf28}+3.09\%$
test_reinforce_speed[reduce-overhead-None] 3.0238ms 2.7900ms 358.4196 Ops/s 355.1204 Ops/s $\color{#35bf28}+0.93\%$
test_reinforce_speed[reduce-overhead-backward] 8.0854ms 7.8549ms 127.3090 Ops/s 122.6448 Ops/s $\color{#35bf28}+3.80\%$
test_iql_speed[False-None] 25.1446ms 19.7461ms 50.6430 Ops/s 50.2618 Ops/s $\color{#35bf28}+0.76\%$
test_iql_speed[False-backward] 35.2724ms 30.0970ms 33.2259 Ops/s 33.4496 Ops/s $\color{#d91a1a}-0.67\%$
test_iql_speed[True-None] 8.7449ms 8.4522ms 118.3121 Ops/s 115.6211 Ops/s $\color{#35bf28}+2.33\%$
test_iql_speed[True-backward] 16.9912ms 16.5043ms 60.5904 Ops/s 60.1239 Ops/s $\color{#35bf28}+0.78\%$
test_iql_speed[reduce-overhead-None] 8.7474ms 8.5254ms 117.2971 Ops/s 117.6008 Ops/s $\color{#d91a1a}-0.26\%$
test_iql_speed[reduce-overhead-backward] 17.2215ms 16.8683ms 59.2826 Ops/s 58.4823 Ops/s $\color{#35bf28}+1.37\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.2935ms 5.8632ms 170.5556 Ops/s 170.4058 Ops/s $\color{#35bf28}+0.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6422ms 0.3601ms 2.7770 KOps/s 2.6443 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6361ms 0.3848ms 2.5990 KOps/s 3.9835 KOps/s $\textbf{\color{#d91a1a}-34.76\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8097ms 5.5462ms 180.3052 Ops/s 178.5302 Ops/s $\color{#35bf28}+0.99\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0133ms 0.3583ms 2.7912 KOps/s 3.1034 KOps/s $\textbf{\color{#d91a1a}-10.06\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5639ms 0.3539ms 2.8260 KOps/s 4.0417 KOps/s $\textbf{\color{#d91a1a}-30.08\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7110ms 1.3947ms 717.0062 Ops/s 815.1113 Ops/s $\textbf{\color{#d91a1a}-12.04\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6542ms 1.4009ms 713.8177 Ops/s 888.2461 Ops/s $\textbf{\color{#d91a1a}-19.64\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.5473ms 5.8642ms 170.5261 Ops/s 173.2181 Ops/s $\color{#d91a1a}-1.55\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9088ms 0.4852ms 2.0612 KOps/s 2.2049 KOps/s $\textbf{\color{#d91a1a}-6.52\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8146ms 0.4723ms 2.1171 KOps/s 2.3521 KOps/s $\textbf{\color{#d91a1a}-9.99\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.6919ms 5.5763ms 179.3313 Ops/s 176.8764 Ops/s $\color{#35bf28}+1.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0034ms 0.3430ms 2.9155 KOps/s 3.1957 KOps/s $\textbf{\color{#d91a1a}-8.77\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5303ms 0.3280ms 3.0488 KOps/s 3.3946 KOps/s $\textbf{\color{#d91a1a}-10.19\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7203ms 5.4948ms 181.9903 Ops/s 178.8000 Ops/s $\color{#35bf28}+1.78\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6877s 1.2763ms 783.5223 Ops/s 3.1429 KOps/s $\textbf{\color{#d91a1a}-75.07\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4489ms 0.2553ms 3.9171 KOps/s 3.7941 KOps/s $\color{#35bf28}+3.24\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8791ms 5.7069ms 175.2267 Ops/s 172.4439 Ops/s $\color{#35bf28}+1.61\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.3989ms 0.4779ms 2.0924 KOps/s 1.9232 KOps/s $\textbf{\color{#35bf28}+8.80\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6977ms 0.4604ms 2.1721 KOps/s 2.0326 KOps/s $\textbf{\color{#35bf28}+6.86\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.5392ms 4.9739ms 201.0479 Ops/s 197.0971 Ops/s $\color{#35bf28}+2.00\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.4986ms 2.0247ms 493.9021 Ops/s 442.9022 Ops/s $\textbf{\color{#35bf28}+11.51\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4256ms 1.1622ms 860.4733 Ops/s 785.2643 Ops/s $\textbf{\color{#35bf28}+9.58\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4889s 14.6424ms 68.2946 Ops/s 196.7353 Ops/s $\textbf{\color{#d91a1a}-65.29\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 10.4910ms 2.1283ms 469.8539 Ops/s 506.9608 Ops/s $\textbf{\color{#d91a1a}-7.32\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.4113ms 1.0755ms 929.7609 Ops/s 827.8119 Ops/s $\textbf{\color{#35bf28}+12.32\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.0597ms 5.2346ms 191.0382 Ops/s 55.3332 Ops/s $\textbf{\color{#35bf28}+245.25\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 7.4584ms 2.1594ms 463.0900 Ops/s 478.7017 Ops/s $\color{#d91a1a}-3.26\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 3.6901ms 1.0268ms 973.8564 Ops/s 1.0333 KOps/s $\textbf{\color{#d91a1a}-5.75\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 34.5236ms 32.1759ms 31.0791 Ops/s 31.2949 Ops/s $\color{#d91a1a}-0.69\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.7646ms 17.0888ms 58.5179 Ops/s 58.8532 Ops/s $\color{#d91a1a}-0.57\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 34.9005ms 32.4776ms 30.7904 Ops/s 30.4476 Ops/s $\color{#35bf28}+1.13\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.4229ms 16.9231ms 59.0908 Ops/s 58.0737 Ops/s $\color{#35bf28}+1.75\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 35.4233ms 33.8247ms 29.5642 Ops/s 28.9066 Ops/s $\color{#35bf28}+2.28\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.7345ms 18.6938ms 53.4937 Ops/s 52.0041 Ops/s $\color{#35bf28}+2.86\%$

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: 8ba7752
Pull-Request: #3211
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: 8ba7752
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
ghstack-source-id: ae4c9e7
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 21, 2025
ghstack-source-id: 03ac637
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 21, 2025
ghstack-source-id: a5e6c7e
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 22, 2025
ghstack-source-id: 2e0ec49
Pull-Request: #3211
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 22, 2025
ghstack-source-id: 6cae8b8
Pull-Request: #3211
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. quality code quality

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant