Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Jul 23, 2025

No description provided.

@pytorch-bot
Copy link

pytorch-bot bot commented Jul 23, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3089

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.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jul 23, 2025
@vmoens vmoens force-pushed the poetry branch 10 times, most recently from fb77ec9 to 1e7e6ea Compare July 24, 2025 02:30
@vmoens vmoens added CI Has to do with CI setup (e.g. wheels & builds, tests...) versioning Versioning change (version number etc) labels Jul 24, 2025
@vmoens vmoens force-pushed the poetry branch 9 times, most recently from 1fdd25c to df508b5 Compare July 25, 2025 11:15
@github-actions
Copy link

github-actions bot commented Jul 25, 2025

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

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

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 84.5067μs 82.7124μs 12.0901 KOps/s 11.9979 KOps/s $\color{#35bf28}+0.77\%$
test_tensor_to_bytestream_speed[torch.save] 0.1437ms 0.1419ms 7.0452 KOps/s 7.0143 KOps/s $\color{#35bf28}+0.44\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1156s 0.1154s 8.6643 Ops/s 8.7153 Ops/s $\color{#d91a1a}-0.59\%$
test_tensor_to_bytestream_speed[numpy] 2.8374μs 2.8319μs 353.1244 KOps/s 359.0223 KOps/s $\color{#d91a1a}-1.64\%$
test_tensor_to_bytestream_speed[safetensors] 42.1231μs 41.9326μs 23.8478 KOps/s 22.8654 KOps/s $\color{#35bf28}+4.30\%$
test_simple 0.5455s 0.5447s 1.8358 Ops/s 1.7669 Ops/s $\color{#35bf28}+3.90\%$
test_transformed 1.1188s 1.1171s 0.8952 Ops/s 0.8828 Ops/s $\color{#35bf28}+1.40\%$
test_serial 1.6854s 1.6814s 0.5947 Ops/s 0.5873 Ops/s $\color{#35bf28}+1.26\%$
test_parallel 1.1562s 1.0818s 0.9244 Ops/s 0.9356 Ops/s $\color{#d91a1a}-1.19\%$
test_step_mdp_speed[True-True-True-True-True] 0.3828ms 46.0719μs 21.7052 KOps/s 22.2722 KOps/s $\color{#d91a1a}-2.55\%$
test_step_mdp_speed[True-True-True-True-False] 55.6710μs 26.3497μs 37.9511 KOps/s 39.8849 KOps/s $\color{#d91a1a}-4.85\%$
test_step_mdp_speed[True-True-True-False-True] 62.7410μs 26.2393μs 38.1107 KOps/s 39.9297 KOps/s $\color{#d91a1a}-4.56\%$
test_step_mdp_speed[True-True-True-False-False] 39.5310μs 14.8050μs 67.5445 KOps/s 71.4562 KOps/s $\textbf{\color{#d91a1a}-5.47\%}$
test_step_mdp_speed[True-True-False-True-True] 86.5110μs 48.5852μs 20.5824 KOps/s 20.6185 KOps/s $\color{#d91a1a}-0.18\%$
test_step_mdp_speed[True-True-False-True-False] 57.0320μs 28.3330μs 35.2945 KOps/s 36.0089 KOps/s $\color{#d91a1a}-1.98\%$
test_step_mdp_speed[True-True-False-False-True] 65.4210μs 29.3999μs 34.0137 KOps/s 35.2429 KOps/s $\color{#d91a1a}-3.49\%$
test_step_mdp_speed[True-True-False-False-False] 50.1510μs 17.2665μs 57.9157 KOps/s 60.9156 KOps/s $\color{#d91a1a}-4.92\%$
test_step_mdp_speed[True-False-True-True-True] 0.1051ms 52.3339μs 19.1081 KOps/s 19.6789 KOps/s $\color{#d91a1a}-2.90\%$
test_step_mdp_speed[True-False-True-True-False] 69.0910μs 31.2505μs 31.9995 KOps/s 32.9682 KOps/s $\color{#d91a1a}-2.94\%$
test_step_mdp_speed[True-False-True-False-True] 60.5310μs 29.2965μs 34.1338 KOps/s 35.4439 KOps/s $\color{#d91a1a}-3.70\%$
test_step_mdp_speed[True-False-True-False-False] 49.7710μs 17.2885μs 57.8420 KOps/s 60.7563 KOps/s $\color{#d91a1a}-4.80\%$
test_step_mdp_speed[True-False-False-True-True] 88.3920μs 53.7689μs 18.5981 KOps/s 18.6477 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[True-False-False-True-False] 80.0420μs 33.6783μs 29.6927 KOps/s 31.1245 KOps/s $\color{#d91a1a}-4.60\%$
test_step_mdp_speed[True-False-False-False-True] 65.7420μs 31.9694μs 31.2800 KOps/s 32.7597 KOps/s $\color{#d91a1a}-4.52\%$
test_step_mdp_speed[True-False-False-False-False] 54.1910μs 20.0133μs 49.9667 KOps/s 51.9738 KOps/s $\color{#d91a1a}-3.86\%$
test_step_mdp_speed[False-True-True-True-True] 91.1320μs 51.7558μs 19.3215 KOps/s 20.0634 KOps/s $\color{#d91a1a}-3.70\%$
test_step_mdp_speed[False-True-True-True-False] 68.5610μs 30.6189μs 32.6595 KOps/s 33.0393 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[False-True-True-False-True] 69.4510μs 32.4813μs 30.7870 KOps/s 31.2236 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[False-True-True-False-False] 53.7820μs 19.3065μs 51.7961 KOps/s 54.2009 KOps/s $\color{#d91a1a}-4.44\%$
test_step_mdp_speed[False-True-False-True-True] 2.7053ms 55.3135μs 18.0788 KOps/s 18.7272 KOps/s $\color{#d91a1a}-3.46\%$
test_step_mdp_speed[False-True-False-True-False] 90.7420μs 33.8162μs 29.5716 KOps/s 30.6158 KOps/s $\color{#d91a1a}-3.41\%$
test_step_mdp_speed[False-True-False-False-True] 68.1920μs 35.7042μs 28.0079 KOps/s 29.3369 KOps/s $\color{#d91a1a}-4.53\%$
test_step_mdp_speed[False-True-False-False-False] 48.3010μs 22.1372μs 45.1729 KOps/s 48.0567 KOps/s $\textbf{\color{#d91a1a}-6.00\%}$
test_step_mdp_speed[False-False-True-True-True] 96.1520μs 57.1368μs 17.5019 KOps/s 17.9918 KOps/s $\color{#d91a1a}-2.72\%$
test_step_mdp_speed[False-False-True-True-False] 68.7520μs 36.3936μs 27.4774 KOps/s 28.3273 KOps/s $\color{#d91a1a}-3.00\%$
test_step_mdp_speed[False-False-True-False-True] 81.0420μs 35.5082μs 28.1625 KOps/s 28.9621 KOps/s $\color{#d91a1a}-2.76\%$
test_step_mdp_speed[False-False-True-False-False] 51.8110μs 21.8263μs 45.8162 KOps/s 48.1825 KOps/s $\color{#d91a1a}-4.91\%$
test_step_mdp_speed[False-False-False-True-True] 0.1027ms 57.8213μs 17.2947 KOps/s 17.3692 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[False-False-False-True-False] 73.7810μs 38.3861μs 26.0511 KOps/s 26.3250 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[False-False-False-False-True] 73.8420μs 37.4974μs 26.6685 KOps/s 26.9575 KOps/s $\color{#d91a1a}-1.07\%$
test_step_mdp_speed[False-False-False-False-False] 59.1210μs 24.4644μs 40.8757 KOps/s 42.9527 KOps/s $\color{#d91a1a}-4.84\%$
test_values[generalized_advantage_estimate-True-True] 11.2389ms 10.9884ms 91.0048 Ops/s 93.0790 Ops/s $\color{#d91a1a}-2.23\%$
test_values[vec_generalized_advantage_estimate-True-True] 20.4810ms 17.3580ms 57.6102 Ops/s 91.1204 Ops/s $\textbf{\color{#d91a1a}-36.78\%}$
test_values[td0_return_estimate-False-False] 0.2445ms 0.1280ms 7.8098 KOps/s 7.7807 KOps/s $\color{#35bf28}+0.37\%$
test_values[td1_return_estimate-False-False] 29.8468ms 28.6999ms 34.8434 Ops/s 35.3826 Ops/s $\color{#d91a1a}-1.52\%$
test_values[vec_td1_return_estimate-False-False] 19.2657ms 17.5856ms 56.8648 Ops/s 90.8542 Ops/s $\textbf{\color{#d91a1a}-37.41\%}$
test_values[td_lambda_return_estimate-True-False] 42.9123ms 41.8797ms 23.8779 Ops/s 23.9344 Ops/s $\color{#d91a1a}-0.24\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.5926ms 17.3385ms 57.6753 Ops/s 90.8480 Ops/s $\textbf{\color{#d91a1a}-36.51\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.8369ms 9.7232ms 102.8467 Ops/s 104.0690 Ops/s $\color{#d91a1a}-1.17\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8622ms 1.5388ms 649.8755 Ops/s 655.2485 Ops/s $\color{#d91a1a}-0.82\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4777ms 0.4241ms 2.3577 KOps/s 2.4199 KOps/s $\color{#d91a1a}-2.57\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.2388ms 33.5219ms 29.8313 Ops/s 34.6218 Ops/s $\textbf{\color{#d91a1a}-13.84\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.1458ms 1.7227ms 580.4896 Ops/s 582.7370 Ops/s $\color{#d91a1a}-0.39\%$
test_dqn_speed[False-None] 6.2262ms 1.3937ms 717.4894 Ops/s 713.0620 Ops/s $\color{#35bf28}+0.62\%$
test_dqn_speed[False-backward] 1.9571ms 1.8925ms 528.4020 Ops/s 525.4587 Ops/s $\color{#35bf28}+0.56\%$
test_dqn_speed[True-None] 0.8954ms 0.5292ms 1.8897 KOps/s 1.8590 KOps/s $\color{#35bf28}+1.65\%$
test_dqn_speed[True-backward] 1.0057ms 0.9696ms 1.0314 KOps/s 908.9571 Ops/s $\textbf{\color{#35bf28}+13.47\%}$
test_dqn_speed[reduce-overhead-None] 0.9085ms 0.5358ms 1.8663 KOps/s 1.8339 KOps/s $\color{#35bf28}+1.77\%$
test_dqn_speed[reduce-overhead-backward] 1.0144ms 0.9772ms 1.0233 KOps/s 1.0213 KOps/s $\color{#35bf28}+0.19\%$
test_ddpg_speed[False-None] 3.2080ms 2.8628ms 349.3107 Ops/s 349.0988 Ops/s $\color{#35bf28}+0.06\%$
test_ddpg_speed[False-backward] 4.2428ms 4.1489ms 241.0297 Ops/s 245.6038 Ops/s $\color{#d91a1a}-1.86\%$
test_ddpg_speed[True-None] 1.7329ms 1.3678ms 731.1166 Ops/s 704.3578 Ops/s $\color{#35bf28}+3.80\%$
test_ddpg_speed[True-backward] 2.3874ms 2.3523ms 425.1216 Ops/s 347.8292 Ops/s $\textbf{\color{#35bf28}+22.22\%}$
test_ddpg_speed[reduce-overhead-None] 1.5660ms 1.3689ms 730.5202 Ops/s 706.9505 Ops/s $\color{#35bf28}+3.33\%$
test_ddpg_speed[reduce-overhead-backward] 2.4336ms 2.3565ms 424.3563 Ops/s 418.0681 Ops/s $\color{#35bf28}+1.50\%$
test_sac_speed[False-None] 8.0876ms 7.6768ms 130.2631 Ops/s 130.4079 Ops/s $\color{#d91a1a}-0.11\%$
test_sac_speed[False-backward] 11.1211ms 10.8081ms 92.5235 Ops/s 92.4436 Ops/s $\color{#35bf28}+0.09\%$
test_sac_speed[True-None] 2.5186ms 2.0759ms 481.7172 Ops/s 471.1726 Ops/s $\color{#35bf28}+2.24\%$
test_sac_speed[True-backward] 4.0351ms 3.9651ms 252.1991 Ops/s 227.8699 Ops/s $\textbf{\color{#35bf28}+10.68\%}$
test_sac_speed[reduce-overhead-None] 2.4539ms 2.0867ms 479.2339 Ops/s 463.7355 Ops/s $\color{#35bf28}+3.34\%$
test_sac_speed[reduce-overhead-backward] 4.0782ms 3.9905ms 250.5979 Ops/s 244.6213 Ops/s $\color{#35bf28}+2.44\%$
test_redq_speed[False-None] 13.2916ms 10.2168ms 97.8783 Ops/s 95.6200 Ops/s $\color{#35bf28}+2.36\%$
test_redq_speed[False-backward] 21.5396ms 17.7812ms 56.2393 Ops/s 54.7986 Ops/s $\color{#35bf28}+2.63\%$
test_redq_speed[True-None] 4.5733ms 4.1770ms 239.4046 Ops/s 229.9266 Ops/s $\color{#35bf28}+4.12\%$
test_redq_speed[True-backward] 9.9002ms 9.5224ms 105.0155 Ops/s 104.3869 Ops/s $\color{#35bf28}+0.60\%$
test_redq_speed[reduce-overhead-None] 4.6454ms 4.2445ms 235.5981 Ops/s 240.3939 Ops/s $\color{#d91a1a}-1.99\%$
test_redq_speed[reduce-overhead-backward] 10.0588ms 9.6698ms 103.4151 Ops/s 103.4113 Ops/s $+0.00\%$
test_redq_deprec_speed[False-None] 11.0782ms 10.6956ms 93.4967 Ops/s 89.8632 Ops/s $\color{#35bf28}+4.04\%$
test_redq_deprec_speed[False-backward] 16.6228ms 15.4857ms 64.5757 Ops/s 61.9666 Ops/s $\color{#35bf28}+4.21\%$
test_redq_deprec_speed[True-None] 3.8546ms 3.5005ms 285.6699 Ops/s 257.8649 Ops/s $\textbf{\color{#35bf28}+10.78\%}$
test_redq_deprec_speed[True-backward] 7.5064ms 7.2631ms 137.6819 Ops/s 125.4640 Ops/s $\textbf{\color{#35bf28}+9.74\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8020ms 3.4450ms 290.2743 Ops/s 259.8723 Ops/s $\textbf{\color{#35bf28}+11.70\%}$
test_redq_deprec_speed[reduce-overhead-backward] 7.6056ms 7.3028ms 136.9340 Ops/s 127.1088 Ops/s $\textbf{\color{#35bf28}+7.73\%}$
test_td3_speed[False-None] 7.8392ms 7.7386ms 129.2219 Ops/s 128.6647 Ops/s $\color{#35bf28}+0.43\%$
test_td3_speed[False-backward] 11.1042ms 10.5360ms 94.9128 Ops/s 94.9191 Ops/s $-0.01\%$
test_td3_speed[True-None] 1.8048ms 1.7554ms 569.6840 Ops/s 549.2278 Ops/s $\color{#35bf28}+3.72\%$
test_td3_speed[True-backward] 4.0440ms 3.5349ms 282.8971 Ops/s 236.0657 Ops/s $\textbf{\color{#35bf28}+19.84\%}$
test_td3_speed[reduce-overhead-None] 1.7840ms 1.7500ms 571.4236 Ops/s 542.2925 Ops/s $\textbf{\color{#35bf28}+5.37\%}$
test_td3_speed[reduce-overhead-backward] 3.6595ms 3.4889ms 286.6194 Ops/s 272.3762 Ops/s $\textbf{\color{#35bf28}+5.23\%}$
test_cql_speed[False-None] 28.8700ms 25.5469ms 39.1436 Ops/s 38.2090 Ops/s $\color{#35bf28}+2.45\%$
test_cql_speed[False-backward] 40.4419ms 35.2941ms 28.3333 Ops/s 28.2517 Ops/s $\color{#35bf28}+0.29\%$
test_cql_speed[True-None] 12.3047ms 11.9856ms 83.4333 Ops/s 83.7443 Ops/s $\color{#d91a1a}-0.37\%$
test_cql_speed[True-backward] 18.2472ms 17.9856ms 55.5999 Ops/s 57.0989 Ops/s $\color{#d91a1a}-2.63\%$
test_cql_speed[reduce-overhead-None] 12.3881ms 12.0205ms 83.1913 Ops/s 82.6743 Ops/s $\color{#35bf28}+0.63\%$
test_cql_speed[reduce-overhead-backward] 19.0081ms 18.1545ms 55.0829 Ops/s 57.0136 Ops/s $\color{#d91a1a}-3.39\%$
test_a2c_speed[False-None] 5.5729ms 5.3363ms 187.3962 Ops/s 184.6011 Ops/s $\color{#35bf28}+1.51\%$
test_a2c_speed[False-backward] 12.2573ms 11.7992ms 84.7517 Ops/s 86.5383 Ops/s $\color{#d91a1a}-2.06\%$
test_a2c_speed[True-None] 3.8409ms 3.6757ms 272.0545 Ops/s 271.5565 Ops/s $\color{#35bf28}+0.18\%$
test_a2c_speed[True-backward] 9.1631ms 8.5324ms 117.2000 Ops/s 117.1439 Ops/s $\color{#35bf28}+0.05\%$
test_a2c_speed[reduce-overhead-None] 3.8546ms 3.6478ms 274.1366 Ops/s 271.6950 Ops/s $\color{#35bf28}+0.90\%$
test_a2c_speed[reduce-overhead-backward] 8.8620ms 8.4892ms 117.7971 Ops/s 118.3036 Ops/s $\color{#d91a1a}-0.43\%$
test_ppo_speed[False-None] 6.2875ms 5.8778ms 170.1325 Ops/s 171.6643 Ops/s $\color{#d91a1a}-0.89\%$
test_ppo_speed[False-backward] 12.8577ms 12.3403ms 81.0353 Ops/s 81.0456 Ops/s $\color{#d91a1a}-0.01\%$
test_ppo_speed[True-None] 3.7330ms 3.5906ms 278.5046 Ops/s 269.7049 Ops/s $\color{#35bf28}+3.26\%$
test_ppo_speed[True-backward] 8.6914ms 8.3818ms 119.3065 Ops/s 94.9756 Ops/s $\textbf{\color{#35bf28}+25.62\%}$
test_ppo_speed[reduce-overhead-None] 4.0250ms 3.6105ms 276.9685 Ops/s 264.0802 Ops/s $\color{#35bf28}+4.88\%$
test_ppo_speed[reduce-overhead-backward] 8.8626ms 8.3930ms 119.1467 Ops/s 119.7163 Ops/s $\color{#d91a1a}-0.48\%$
test_reinforce_speed[False-None] 4.8227ms 4.5314ms 220.6844 Ops/s 222.2728 Ops/s $\color{#d91a1a}-0.71\%$
test_reinforce_speed[False-backward] 7.4589ms 7.2535ms 137.8648 Ops/s 137.2265 Ops/s $\color{#35bf28}+0.47\%$
test_reinforce_speed[True-None] 3.2225ms 2.8617ms 349.4467 Ops/s 341.2515 Ops/s $\color{#35bf28}+2.40\%$
test_reinforce_speed[True-backward] 7.7499ms 7.5027ms 133.2860 Ops/s 120.1104 Ops/s $\textbf{\color{#35bf28}+10.97\%}$
test_reinforce_speed[reduce-overhead-None] 3.3566ms 2.8426ms 351.7872 Ops/s 351.2063 Ops/s $\color{#35bf28}+0.17\%$
test_reinforce_speed[reduce-overhead-backward] 7.7681ms 7.5222ms 132.9401 Ops/s 129.4026 Ops/s $\color{#35bf28}+2.73\%$
test_iql_speed[False-None] 25.3236ms 20.2167ms 49.4641 Ops/s 49.7710 Ops/s $\color{#d91a1a}-0.62\%$
test_iql_speed[False-backward] 36.1457ms 30.8223ms 32.4440 Ops/s 33.4317 Ops/s $\color{#d91a1a}-2.95\%$
test_iql_speed[True-None] 8.9876ms 8.3664ms 119.5262 Ops/s 116.3648 Ops/s $\color{#35bf28}+2.72\%$
test_iql_speed[True-backward] 16.9756ms 16.5511ms 60.4191 Ops/s 59.9897 Ops/s $\color{#35bf28}+0.72\%$
test_iql_speed[reduce-overhead-None] 8.6809ms 8.4066ms 118.9547 Ops/s 119.6920 Ops/s $\color{#d91a1a}-0.62\%$
test_iql_speed[reduce-overhead-backward] 16.8639ms 16.4350ms 60.8459 Ops/s 61.0624 Ops/s $\color{#d91a1a}-0.35\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5527ms 6.1967ms 161.3762 Ops/s 163.1904 Ops/s $\color{#d91a1a}-1.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6007ms 0.3003ms 3.3295 KOps/s 3.7714 KOps/s $\textbf{\color{#d91a1a}-11.72\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6456ms 0.2824ms 3.5408 KOps/s 4.1040 KOps/s $\textbf{\color{#d91a1a}-13.72\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1983ms 5.9240ms 168.8041 Ops/s 169.6083 Ops/s $\color{#d91a1a}-0.47\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7286ms 0.2903ms 3.4450 KOps/s 3.8676 KOps/s $\textbf{\color{#d91a1a}-10.92\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5012ms 0.2664ms 3.7539 KOps/s 4.1760 KOps/s $\textbf{\color{#d91a1a}-10.11\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5640ms 1.3080ms 764.5227 Ops/s 761.7097 Ops/s $\color{#35bf28}+0.37\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5287ms 1.2271ms 814.9448 Ops/s 882.9915 Ops/s $\textbf{\color{#d91a1a}-7.71\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3089ms 6.0799ms 164.4766 Ops/s 165.3001 Ops/s $\color{#d91a1a}-0.50\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8789ms 0.4344ms 2.3020 KOps/s 2.4376 KOps/s $\textbf{\color{#d91a1a}-5.57\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6907ms 0.4156ms 2.4064 KOps/s 2.6006 KOps/s $\textbf{\color{#d91a1a}-7.47\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0749ms 5.9679ms 167.5622 Ops/s 167.5327 Ops/s $\color{#35bf28}+0.02\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0136ms 0.3142ms 3.1827 KOps/s 2.9232 KOps/s $\textbf{\color{#35bf28}+8.88\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7029ms 0.2876ms 3.4770 KOps/s 4.1093 KOps/s $\textbf{\color{#d91a1a}-15.39\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 10.1097ms 5.9307ms 168.6127 Ops/s 169.0466 Ops/s $\color{#d91a1a}-0.26\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.4613ms 0.3698ms 2.7045 KOps/s 3.1467 KOps/s $\textbf{\color{#d91a1a}-14.05\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5808ms 0.2868ms 3.4864 KOps/s 3.5063 KOps/s $\color{#d91a1a}-0.57\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2789ms 6.1090ms 163.6936 Ops/s 163.6168 Ops/s $\color{#35bf28}+0.05\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8271ms 0.4982ms 2.0072 KOps/s 2.0860 KOps/s $\color{#d91a1a}-3.78\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7218ms 0.4832ms 2.0696 KOps/s 2.1683 KOps/s $\color{#d91a1a}-4.55\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.3214ms 5.7185ms 174.8725 Ops/s 178.3655 Ops/s $\color{#d91a1a}-1.96\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.2738ms 2.1222ms 471.1996 Ops/s 404.8173 Ops/s $\textbf{\color{#35bf28}+16.40\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.9778ms 1.2448ms 803.3679 Ops/s 842.2292 Ops/s $\color{#d91a1a}-4.61\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4191s 14.0382ms 71.2340 Ops/s 59.9319 Ops/s $\textbf{\color{#35bf28}+18.86\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.7818ms 2.0585ms 485.7878 Ops/s 538.5161 Ops/s $\textbf{\color{#d91a1a}-9.79\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.1714ms 0.9294ms 1.0760 KOps/s 758.2951 Ops/s $\textbf{\color{#35bf28}+41.89\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.5860ms 5.9907ms 166.9240 Ops/s 172.4252 Ops/s $\color{#d91a1a}-3.19\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.3635ms 2.2505ms 444.3361 Ops/s 439.6730 Ops/s $\color{#35bf28}+1.06\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 3.6589ms 1.3241ms 755.2095 Ops/s 723.2032 Ops/s $\color{#35bf28}+4.43\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 62.9191ms 58.6664ms 17.0455 Ops/s 16.8390 Ops/s $\color{#35bf28}+1.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.1665ms 16.6929ms 59.9056 Ops/s 60.3930 Ops/s $\color{#d91a1a}-0.81\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 62.9060ms 60.1536ms 16.6241 Ops/s 16.9944 Ops/s $\color{#d91a1a}-2.18\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.9201ms 17.1713ms 58.2368 Ops/s 59.1625 Ops/s $\color{#d91a1a}-1.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 60.5762ms 58.3792ms 17.1294 Ops/s 16.5187 Ops/s $\color{#35bf28}+3.70\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.2733ms 18.2728ms 54.7261 Ops/s 54.6509 Ops/s $\color{#35bf28}+0.14\%$

@github-actions
Copy link

github-actions bot commented Jul 25, 2025

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

Total Benchmarks: 148. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 87.1061μs 84.8536μs 11.7850 KOps/s 12.1706 KOps/s $\color{#d91a1a}-3.17\%$
test_tensor_to_bytestream_speed[torch.save] 0.1500ms 0.1482ms 6.7494 KOps/s 6.9147 KOps/s $\color{#d91a1a}-2.39\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1170s 0.1164s 8.5877 Ops/s 8.5552 Ops/s $\color{#35bf28}+0.38\%$
test_tensor_to_bytestream_speed[numpy] 2.8435μs 2.8370μs 352.4821 KOps/s 359.3014 KOps/s $\color{#d91a1a}-1.90\%$
test_tensor_to_bytestream_speed[safetensors] 43.5487μs 43.2893μs 23.1004 KOps/s 23.5098 KOps/s $\color{#d91a1a}-1.74\%$
test_simple 0.7973s 0.7852s 1.2735 Ops/s 1.2396 Ops/s $\color{#35bf28}+2.74\%$
test_transformed 1.4117s 1.4066s 0.7110 Ops/s 0.7000 Ops/s $\color{#35bf28}+1.56\%$
test_serial 2.3056s 2.2955s 0.4356 Ops/s 0.4243 Ops/s $\color{#35bf28}+2.68\%$
test_parallel 2.0054s 1.9241s 0.5197 Ops/s 0.5304 Ops/s $\color{#d91a1a}-2.01\%$
test_step_mdp_speed[True-True-True-True-True] 0.2068ms 44.5764μs 22.4334 KOps/s 22.0274 KOps/s $\color{#35bf28}+1.84\%$
test_step_mdp_speed[True-True-True-True-False] 0.1160ms 25.6426μs 38.9976 KOps/s 39.4507 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[True-True-True-False-True] 0.1160ms 25.0236μs 39.9622 KOps/s 39.2729 KOps/s $\color{#35bf28}+1.76\%$
test_step_mdp_speed[True-True-True-False-False] 81.9410μs 14.4076μs 69.4080 KOps/s 72.2802 KOps/s $\color{#d91a1a}-3.97\%$
test_step_mdp_speed[True-True-False-True-True] 89.8710μs 47.4594μs 21.0706 KOps/s 20.9060 KOps/s $\color{#35bf28}+0.79\%$
test_step_mdp_speed[True-True-False-True-False] 67.9410μs 27.9456μs 35.7838 KOps/s 35.0095 KOps/s $\color{#35bf28}+2.21\%$
test_step_mdp_speed[True-True-False-False-True] 63.4910μs 27.8472μs 35.9102 KOps/s 35.5620 KOps/s $\color{#35bf28}+0.98\%$
test_step_mdp_speed[True-True-False-False-False] 52.2810μs 17.0524μs 58.6429 KOps/s 59.3895 KOps/s $\color{#d91a1a}-1.26\%$
test_step_mdp_speed[True-False-True-True-True] 92.0310μs 50.1202μs 19.9520 KOps/s 19.6717 KOps/s $\color{#35bf28}+1.43\%$
test_step_mdp_speed[True-False-True-True-False] 94.4610μs 30.5474μs 32.7360 KOps/s 31.9531 KOps/s $\color{#35bf28}+2.45\%$
test_step_mdp_speed[True-False-True-False-True] 63.4010μs 27.7666μs 36.0145 KOps/s 35.4776 KOps/s $\color{#35bf28}+1.51\%$
test_step_mdp_speed[True-False-True-False-False] 56.7310μs 16.7642μs 59.6509 KOps/s 59.3777 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[True-False-False-True-True] 0.1108ms 52.8444μs 18.9235 KOps/s 18.5466 KOps/s $\color{#35bf28}+2.03\%$
test_step_mdp_speed[True-False-False-True-False] 64.9410μs 33.5014μs 29.8495 KOps/s 29.2699 KOps/s $\color{#35bf28}+1.98\%$
test_step_mdp_speed[True-False-False-False-True] 63.1110μs 30.4562μs 32.8341 KOps/s 32.0737 KOps/s $\color{#35bf28}+2.37\%$
test_step_mdp_speed[True-False-False-False-False] 54.2110μs 19.6416μs 50.9124 KOps/s 51.0528 KOps/s $\color{#d91a1a}-0.28\%$
test_step_mdp_speed[False-True-True-True-True] 94.8710μs 49.6797μs 20.1290 KOps/s 19.3944 KOps/s $\color{#35bf28}+3.79\%$
test_step_mdp_speed[False-True-True-True-False] 69.1310μs 30.7512μs 32.5191 KOps/s 32.1944 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[False-True-True-False-True] 88.6710μs 31.3620μs 31.8857 KOps/s 31.7813 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[False-True-True-False-False] 8.7722ms 19.0096μs 52.6050 KOps/s 52.3184 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-True-False-True-True] 2.6641ms 53.4569μs 18.7067 KOps/s 18.5700 KOps/s $\color{#35bf28}+0.74\%$
test_step_mdp_speed[False-True-False-True-False] 72.1520μs 33.6527μs 29.7153 KOps/s 29.2456 KOps/s $\color{#35bf28}+1.61\%$
test_step_mdp_speed[False-True-False-False-True] 0.1109ms 34.5051μs 28.9812 KOps/s 28.9212 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[False-True-False-False-False] 51.8310μs 21.4404μs 46.6409 KOps/s 46.0335 KOps/s $\color{#35bf28}+1.32\%$
test_step_mdp_speed[False-False-True-True-True] 0.2312ms 55.5935μs 17.9877 KOps/s 17.8822 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[False-False-True-True-False] 72.9710μs 35.9222μs 27.8379 KOps/s 27.0110 KOps/s $\color{#35bf28}+3.06\%$
test_step_mdp_speed[False-False-True-False-True] 75.5210μs 34.3479μs 29.1139 KOps/s 29.0077 KOps/s $\color{#35bf28}+0.37\%$
test_step_mdp_speed[False-False-True-False-False] 0.2170ms 21.3029μs 46.9420 KOps/s 45.7461 KOps/s $\color{#35bf28}+2.61\%$
test_step_mdp_speed[False-False-False-True-True] 0.1004ms 58.3104μs 17.1496 KOps/s 16.8924 KOps/s $\color{#35bf28}+1.52\%$
test_step_mdp_speed[False-False-False-True-False] 78.4420μs 38.0690μs 26.2681 KOps/s 24.9564 KOps/s $\textbf{\color{#35bf28}+5.26\%}$
test_step_mdp_speed[False-False-False-False-True] 0.2195ms 36.5401μs 27.3672 KOps/s 26.9935 KOps/s $\color{#35bf28}+1.38\%$
test_step_mdp_speed[False-False-False-False-False] 81.4620μs 23.7234μs 42.1525 KOps/s 41.1396 KOps/s $\color{#35bf28}+2.46\%$
test_values[generalized_advantage_estimate-True-True] 23.1110ms 22.1869ms 45.0716 Ops/s 44.1241 Ops/s $\color{#35bf28}+2.15\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1255s 3.4341ms 291.1950 Ops/s 285.2625 Ops/s $\color{#35bf28}+2.08\%$
test_values[td0_return_estimate-False-False] 0.1060ms 81.2155μs 12.3129 KOps/s 12.2697 KOps/s $\color{#35bf28}+0.35\%$
test_values[td1_return_estimate-False-False] 53.6917ms 52.4640ms 19.0607 Ops/s 18.5905 Ops/s $\color{#35bf28}+2.53\%$
test_values[vec_td1_return_estimate-False-False] 1.3983ms 1.1140ms 897.6753 Ops/s 892.1559 Ops/s $\color{#35bf28}+0.62\%$
test_values[td_lambda_return_estimate-True-False] 87.0878ms 83.6380ms 11.9563 Ops/s 11.4353 Ops/s $\color{#35bf28}+4.56\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3474ms 1.1086ms 902.0479 Ops/s 896.1119 Ops/s $\color{#35bf28}+0.66\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 22.3637ms 22.0220ms 45.4092 Ops/s 42.7975 Ops/s $\textbf{\color{#35bf28}+6.10\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0403ms 0.7676ms 1.3027 KOps/s 1.3025 KOps/s $\color{#35bf28}+0.01\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8503ms 0.6844ms 1.4611 KOps/s 1.4522 KOps/s $\color{#35bf28}+0.61\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6807ms 1.4977ms 667.6816 Ops/s 665.5578 Ops/s $\color{#35bf28}+0.32\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8834ms 0.6985ms 1.4316 KOps/s 1.4233 KOps/s $\color{#35bf28}+0.58\%$
test_dqn_speed[False-None] 6.6408ms 1.5753ms 634.8184 Ops/s 627.6912 Ops/s $\color{#35bf28}+1.14\%$
test_dqn_speed[False-backward] 2.4158ms 2.2274ms 448.9587 Ops/s 436.8533 Ops/s $\color{#35bf28}+2.77\%$
test_dqn_speed[True-None] 0.7658ms 0.5932ms 1.6858 KOps/s 1.6710 KOps/s $\color{#35bf28}+0.88\%$
test_dqn_speed[True-backward] 1.4567ms 1.3099ms 763.3897 Ops/s 752.3464 Ops/s $\color{#35bf28}+1.47\%$
test_dqn_speed[reduce-overhead-None] 0.7628ms 0.6107ms 1.6375 KOps/s 1.6265 KOps/s $\color{#35bf28}+0.68\%$
test_dqn_speed[reduce-overhead-backward] 1.1691ms 1.1313ms 883.9551 Ops/s 874.4141 Ops/s $\color{#35bf28}+1.09\%$
test_ddpg_speed[False-None] 3.3288ms 2.9889ms 334.5677 Ops/s 334.4107 Ops/s $\color{#35bf28}+0.05\%$
test_ddpg_speed[False-backward] 4.8132ms 4.4639ms 224.0207 Ops/s 221.9920 Ops/s $\color{#35bf28}+0.91\%$
test_ddpg_speed[True-None] 1.7959ms 1.4038ms 712.3555 Ops/s 711.2766 Ops/s $\color{#35bf28}+0.15\%$
test_ddpg_speed[True-backward] 2.9495ms 2.7987ms 357.3062 Ops/s 352.4901 Ops/s $\color{#35bf28}+1.37\%$
test_ddpg_speed[reduce-overhead-None] 1.8235ms 1.4342ms 697.2305 Ops/s 700.1504 Ops/s $\color{#d91a1a}-0.42\%$
test_ddpg_speed[reduce-overhead-backward] 0.1878s 0.1853s 5.3968 Ops/s 4.4528 Ops/s $\textbf{\color{#35bf28}+21.20\%}$
test_sac_speed[False-None] 8.8726ms 8.2597ms 121.0700 Ops/s 120.2745 Ops/s $\color{#35bf28}+0.66\%$
test_sac_speed[False-backward] 12.2037ms 11.6193ms 86.0640 Ops/s 84.5948 Ops/s $\color{#35bf28}+1.74\%$
test_sac_speed[True-None] 2.3894ms 1.9310ms 517.8574 Ops/s 514.7939 Ops/s $\color{#35bf28}+0.60\%$
test_sac_speed[True-backward] 4.3672ms 4.0065ms 249.5930 Ops/s 248.2935 Ops/s $\color{#35bf28}+0.52\%$
test_sac_speed[reduce-overhead-None] 20.3593ms 11.5850ms 86.3182 Ops/s 87.0506 Ops/s $\color{#d91a1a}-0.84\%$
test_sac_speed[reduce-overhead-backward] 1.9520ms 1.7971ms 556.4400 Ops/s 555.4661 Ops/s $\color{#35bf28}+0.18\%$
test_redq_deprec_speed[False-None] 9.6423ms 9.1569ms 109.2075 Ops/s 106.7684 Ops/s $\color{#35bf28}+2.28\%$
test_redq_deprec_speed[False-backward] 13.3798ms 12.7755ms 78.2750 Ops/s 77.1275 Ops/s $\color{#35bf28}+1.49\%$
test_redq_deprec_speed[True-None] 3.1160ms 2.6551ms 376.6293 Ops/s 383.8612 Ops/s $\color{#d91a1a}-1.88\%$
test_redq_deprec_speed[True-backward] 5.2089ms 4.6268ms 216.1343 Ops/s 214.8231 Ops/s $\color{#35bf28}+0.61\%$
test_redq_deprec_speed[reduce-overhead-None] 3.0072ms 2.5752ms 388.3123 Ops/s 381.6657 Ops/s $\color{#35bf28}+1.74\%$
test_redq_deprec_speed[reduce-overhead-backward] 5.0695ms 4.6429ms 215.3841 Ops/s 210.6972 Ops/s $\color{#35bf28}+2.22\%$
test_td3_speed[False-None] 8.3408ms 8.1451ms 122.7729 Ops/s 121.5361 Ops/s $\color{#35bf28}+1.02\%$
test_td3_speed[False-backward] 11.6632ms 10.9555ms 91.2784 Ops/s 89.9252 Ops/s $\color{#35bf28}+1.50\%$
test_td3_speed[True-None] 1.7933ms 1.7442ms 573.3354 Ops/s 558.4914 Ops/s $\color{#35bf28}+2.66\%$
test_td3_speed[True-backward] 4.0146ms 3.6224ms 276.0589 Ops/s 273.1302 Ops/s $\color{#35bf28}+1.07\%$
test_td3_speed[reduce-overhead-None] 50.7179ms 25.8804ms 38.6393 Ops/s 37.5444 Ops/s $\color{#35bf28}+2.92\%$
test_td3_speed[reduce-overhead-backward] 1.4542ms 1.3188ms 758.2515 Ops/s 671.4688 Ops/s $\textbf{\color{#35bf28}+12.92\%}$
test_cql_speed[False-None] 18.1908ms 17.4029ms 57.4618 Ops/s 57.5447 Ops/s $\color{#d91a1a}-0.14\%$
test_cql_speed[False-backward] 23.8819ms 22.9935ms 43.4905 Ops/s 42.9110 Ops/s $\color{#35bf28}+1.35\%$
test_cql_speed[True-None] 3.8941ms 3.5503ms 281.6678 Ops/s 281.1922 Ops/s $\color{#35bf28}+0.17\%$
test_cql_speed[True-backward] 6.3085ms 6.1748ms 161.9486 Ops/s 160.6129 Ops/s $\color{#35bf28}+0.83\%$
test_cql_speed[reduce-overhead-None] 20.0823ms 12.6602ms 78.9879 Ops/s 76.9031 Ops/s $\color{#35bf28}+2.71\%$
test_cql_speed[reduce-overhead-backward] 1.9705ms 1.8356ms 544.7901 Ops/s 558.3932 Ops/s $\color{#d91a1a}-2.44\%$
test_a2c_speed[False-None] 3.5843ms 3.2979ms 303.2206 Ops/s 303.9633 Ops/s $\color{#d91a1a}-0.24\%$
test_a2c_speed[False-backward] 7.2189ms 6.5953ms 151.6241 Ops/s 153.8830 Ops/s $\color{#d91a1a}-1.47\%$
test_a2c_speed[True-None] 1.5151ms 1.3655ms 732.3098 Ops/s 737.0763 Ops/s $\color{#d91a1a}-0.65\%$
test_a2c_speed[True-backward] 3.3996ms 3.2622ms 306.5432 Ops/s 315.7408 Ops/s $\color{#d91a1a}-2.91\%$
test_a2c_speed[reduce-overhead-None] 15.5709ms 8.8630ms 112.8292 Ops/s 110.5927 Ops/s $\color{#35bf28}+2.02\%$
test_a2c_speed[reduce-overhead-backward] 1.6451ms 1.5141ms 660.4612 Ops/s 699.3856 Ops/s $\textbf{\color{#d91a1a}-5.57\%}$
test_ppo_speed[False-None] 4.0214ms 3.8661ms 258.6567 Ops/s 255.7194 Ops/s $\color{#35bf28}+1.15\%$
test_ppo_speed[False-backward] 7.4662ms 7.3178ms 136.6540 Ops/s 137.7541 Ops/s $\color{#d91a1a}-0.80\%$
test_ppo_speed[True-None] 1.6475ms 1.4713ms 679.6720 Ops/s 663.2436 Ops/s $\color{#35bf28}+2.48\%$
test_ppo_speed[True-backward] 3.7932ms 3.3244ms 300.8060 Ops/s 287.0143 Ops/s $\color{#35bf28}+4.81\%$
test_ppo_speed[reduce-overhead-None] 1.6632ms 1.4512ms 689.0928 Ops/s 677.7079 Ops/s $\color{#35bf28}+1.68\%$
test_ppo_speed[reduce-overhead-backward] 3.6345ms 3.2981ms 303.2050 Ops/s 287.7311 Ops/s $\textbf{\color{#35bf28}+5.38\%}$
test_reinforce_speed[False-None] 2.5098ms 2.3415ms 427.0841 Ops/s 421.8438 Ops/s $\color{#35bf28}+1.24\%$
test_reinforce_speed[False-backward] 3.5472ms 3.3962ms 294.4492 Ops/s 283.4965 Ops/s $\color{#35bf28}+3.86\%$
test_reinforce_speed[True-None] 1.5926ms 1.3286ms 752.6723 Ops/s 753.5537 Ops/s $\color{#d91a1a}-0.12\%$
test_reinforce_speed[True-backward] 3.2839ms 3.1252ms 319.9755 Ops/s 315.1220 Ops/s $\color{#35bf28}+1.54\%$
test_reinforce_speed[reduce-overhead-None] 19.1536ms 10.4520ms 95.6757 Ops/s 93.7903 Ops/s $\color{#35bf28}+2.01\%$
test_reinforce_speed[reduce-overhead-backward] 1.6844ms 1.5016ms 665.9549 Ops/s 664.0499 Ops/s $\color{#35bf28}+0.29\%$
test_iql_speed[False-None] 10.1492ms 9.5221ms 105.0184 Ops/s 104.3813 Ops/s $\color{#35bf28}+0.61\%$
test_iql_speed[False-backward] 14.1540ms 13.5553ms 73.7721 Ops/s 73.6012 Ops/s $\color{#35bf28}+0.23\%$
test_iql_speed[True-None] 2.5302ms 2.3561ms 424.4244 Ops/s 424.9568 Ops/s $\color{#d91a1a}-0.13\%$
test_iql_speed[True-backward] 5.3036ms 5.1487ms 194.2254 Ops/s 189.6346 Ops/s $\color{#35bf28}+2.42\%$
test_iql_speed[reduce-overhead-None] 18.3761ms 10.8346ms 92.2970 Ops/s 90.7378 Ops/s $\color{#35bf28}+1.72\%$
test_iql_speed[reduce-overhead-backward] 2.0504ms 1.9092ms 523.7858 Ops/s 502.3252 Ops/s $\color{#35bf28}+4.27\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7284ms 6.2788ms 159.2669 Ops/s 158.1648 Ops/s $\color{#35bf28}+0.70\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6192ms 0.3312ms 3.0189 KOps/s 3.1266 KOps/s $\color{#d91a1a}-3.44\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5835ms 0.3251ms 3.0760 KOps/s 3.2109 KOps/s $\color{#d91a1a}-4.20\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3294ms 6.0291ms 165.8616 Ops/s 165.1464 Ops/s $\color{#35bf28}+0.43\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7898ms 0.3314ms 3.0174 KOps/s 3.5236 KOps/s $\textbf{\color{#d91a1a}-14.36\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6077ms 0.3152ms 3.1723 KOps/s 3.4419 KOps/s $\textbf{\color{#d91a1a}-7.83\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6827ms 1.3798ms 724.7308 Ops/s 778.7835 Ops/s $\textbf{\color{#d91a1a}-6.94\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6069ms 1.3160ms 759.8586 Ops/s 756.7074 Ops/s $\color{#35bf28}+0.42\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4096ms 6.1837ms 161.7154 Ops/s 161.3078 Ops/s $\color{#35bf28}+0.25\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9409ms 0.4717ms 2.1200 KOps/s 2.2248 KOps/s $\color{#d91a1a}-4.71\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9513ms 0.4551ms 2.1973 KOps/s 2.4268 KOps/s $\textbf{\color{#d91a1a}-9.46\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1515ms 5.9297ms 168.6437 Ops/s 166.2231 Ops/s $\color{#35bf28}+1.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9234ms 0.3163ms 3.1615 KOps/s 2.9760 KOps/s $\textbf{\color{#35bf28}+6.24\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5534ms 0.2545ms 3.9299 KOps/s 3.2997 KOps/s $\textbf{\color{#35bf28}+19.10\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2362ms 5.9201ms 168.9150 Ops/s 166.9834 Ops/s $\color{#35bf28}+1.16\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9440ms 0.3049ms 3.2800 KOps/s 3.6771 KOps/s $\textbf{\color{#d91a1a}-10.80\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6449ms 0.2999ms 3.3348 KOps/s 3.4340 KOps/s $\color{#d91a1a}-2.89\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3850ms 6.1122ms 163.6065 Ops/s 162.3203 Ops/s $\color{#35bf28}+0.79\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1441ms 0.4579ms 2.1841 KOps/s 2.2222 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7025ms 0.4398ms 2.2738 KOps/s 2.1331 KOps/s $\textbf{\color{#35bf28}+6.59\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.3833ms 5.6502ms 176.9848 Ops/s 52.2243 Ops/s $\textbf{\color{#35bf28}+238.89\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.8703ms 2.2267ms 449.0961 Ops/s 465.7640 Ops/s $\color{#d91a1a}-3.58\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4295ms 1.3473ms 742.2205 Ops/s 800.9490 Ops/s $\textbf{\color{#d91a1a}-7.33\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.3994ms 5.7094ms 175.1505 Ops/s 177.8392 Ops/s $\color{#d91a1a}-1.51\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 6.1592ms 2.1577ms 463.4514 Ops/s 482.2007 Ops/s $\color{#d91a1a}-3.89\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 10.5372ms 1.4119ms 708.2604 Ops/s 763.5710 Ops/s $\textbf{\color{#d91a1a}-7.24\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4985s 15.8164ms 63.2255 Ops/s 166.9927 Ops/s $\textbf{\color{#d91a1a}-62.14\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.5795ms 2.2916ms 436.3851 Ops/s 439.8718 Ops/s $\color{#d91a1a}-0.79\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.4658ms 1.2781ms 782.4364 Ops/s 692.3689 Ops/s $\textbf{\color{#35bf28}+13.01\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 63.7951ms 60.6842ms 16.4788 Ops/s 17.1745 Ops/s $\color{#d91a1a}-4.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.8677ms 17.2768ms 57.8811 Ops/s 59.8881 Ops/s $\color{#d91a1a}-3.35\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 63.1539ms 60.1866ms 16.6150 Ops/s 17.1488 Ops/s $\color{#d91a1a}-3.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.4009ms 17.3230ms 57.7267 Ops/s 58.6775 Ops/s $\color{#d91a1a}-1.62\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 63.7093ms 60.6142ms 16.4978 Ops/s 16.9746 Ops/s $\color{#d91a1a}-2.81\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.0722ms 18.6121ms 53.7284 Ops/s 53.9705 Ops/s $\color{#d91a1a}-0.45\%$

@vmoens vmoens force-pushed the poetry branch 2 times, most recently from d15bffc to 64f09b7 Compare July 28, 2025 20:58
@vmoens vmoens force-pushed the poetry branch 8 times, most recently from 6e87eaf to c5a2e89 Compare July 29, 2025 10:12
@vmoens vmoens force-pushed the poetry branch 9 times, most recently from 21469c0 to a58f391 Compare July 29, 2025 15:34
@vmoens vmoens merged commit 0c05bba into main Jul 29, 2025
53 of 72 checks passed
@vmoens vmoens deleted the poetry branch July 29, 2025 15:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CI Has to do with CI setup (e.g. wheels & builds, tests...) CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. versioning Versioning change (version number etc)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants