Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Deprecation] Remove InPlaceSampler #2750

Merged
merged 3 commits into from
Feb 4, 2025
Merged

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 3, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2750

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

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

[ghstack-poisoned]
Copy link

github-actions bot commented Feb 3, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5258s 0.4449s 2.2478 Ops/s 2.2205 Ops/s $\color{#35bf28}+1.23\%$
test_transformed 0.9896s 0.9080s 1.1013 Ops/s 1.0822 Ops/s $\color{#35bf28}+1.77\%$
test_serial 1.4491s 1.3661s 0.7320 Ops/s 0.7225 Ops/s $\color{#35bf28}+1.32\%$
test_parallel 1.2938s 1.2025s 0.8316 Ops/s 0.8131 Ops/s $\color{#35bf28}+2.27\%$
test_step_mdp_speed[True-True-True-True-True] 0.3181ms 30.1756μs 33.1393 KOps/s 33.6438 KOps/s $\color{#d91a1a}-1.50\%$
test_step_mdp_speed[True-True-True-True-False] 54.7820μs 18.4476μs 54.2076 KOps/s 57.7391 KOps/s $\textbf{\color{#d91a1a}-6.12\%}$
test_step_mdp_speed[True-True-True-False-True] 49.7320μs 17.3512μs 57.6329 KOps/s 59.7812 KOps/s $\color{#d91a1a}-3.59\%$
test_step_mdp_speed[True-True-True-False-False] 36.0970μs 10.2503μs 97.5584 KOps/s 102.5689 KOps/s $\color{#d91a1a}-4.88\%$
test_step_mdp_speed[True-True-False-True-True] 76.0620μs 32.2364μs 31.0208 KOps/s 31.5472 KOps/s $\color{#d91a1a}-1.67\%$
test_step_mdp_speed[True-True-False-True-False] 49.1320μs 20.0571μs 49.8577 KOps/s 52.3175 KOps/s $\color{#d91a1a}-4.70\%$
test_step_mdp_speed[True-True-False-False-True] 59.0800μs 19.0827μs 52.4036 KOps/s 54.0218 KOps/s $\color{#d91a1a}-3.00\%$
test_step_mdp_speed[True-True-False-False-False] 34.9150μs 12.1557μs 82.2663 KOps/s 85.2610 KOps/s $\color{#d91a1a}-3.51\%$
test_step_mdp_speed[True-False-True-True-True] 72.1350μs 34.3561μs 29.1069 KOps/s 29.6133 KOps/s $\color{#d91a1a}-1.71\%$
test_step_mdp_speed[True-False-True-True-False] 62.7170μs 21.8374μs 45.7929 KOps/s 47.3615 KOps/s $\color{#d91a1a}-3.31\%$
test_step_mdp_speed[True-False-True-False-True] 45.6250μs 19.0802μs 52.4105 KOps/s 52.7452 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[True-False-True-False-False] 35.4460μs 12.2762μs 81.4584 KOps/s 85.6951 KOps/s $\color{#d91a1a}-4.94\%$
test_step_mdp_speed[True-False-False-True-True] 87.0520μs 35.9517μs 27.8151 KOps/s 27.8994 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-False-False-True-False] 57.5070μs 23.8611μs 41.9092 KOps/s 43.6401 KOps/s $\color{#d91a1a}-3.97\%$
test_step_mdp_speed[True-False-False-False-True] 48.9210μs 21.0502μs 47.5054 KOps/s 48.6473 KOps/s $\color{#d91a1a}-2.35\%$
test_step_mdp_speed[True-False-False-False-False] 45.3250μs 14.1345μs 70.7487 KOps/s 74.9743 KOps/s $\textbf{\color{#d91a1a}-5.64\%}$
test_step_mdp_speed[False-True-True-True-True] 78.5260μs 34.3376μs 29.1226 KOps/s 29.6969 KOps/s $\color{#d91a1a}-1.93\%$
test_step_mdp_speed[False-True-True-True-False] 56.6550μs 22.1389μs 45.1694 KOps/s 47.2584 KOps/s $\color{#d91a1a}-4.42\%$
test_step_mdp_speed[False-True-True-False-True] 68.1270μs 21.9383μs 45.5824 KOps/s 46.5279 KOps/s $\color{#d91a1a}-2.03\%$
test_step_mdp_speed[False-True-True-False-False] 35.6760μs 13.7886μs 72.5235 KOps/s 76.8415 KOps/s $\textbf{\color{#d91a1a}-5.62\%}$
test_step_mdp_speed[False-True-False-True-True] 84.2770μs 35.9444μs 27.8208 KOps/s 28.4006 KOps/s $\color{#d91a1a}-2.04\%$
test_step_mdp_speed[False-True-False-True-False] 62.0660μs 24.0751μs 41.5366 KOps/s 43.6503 KOps/s $\color{#d91a1a}-4.84\%$
test_step_mdp_speed[False-True-False-False-True] 2.7789ms 24.1383μs 41.4279 KOps/s 42.8807 KOps/s $\color{#d91a1a}-3.39\%$
test_step_mdp_speed[False-True-False-False-False] 53.8800μs 15.5165μs 64.4474 KOps/s 68.0474 KOps/s $\textbf{\color{#d91a1a}-5.29\%}$
test_step_mdp_speed[False-False-True-True-True] 82.3030μs 38.0865μs 26.2560 KOps/s 26.9894 KOps/s $\color{#d91a1a}-2.72\%$
test_step_mdp_speed[False-False-True-True-False] 70.3610μs 25.8842μs 38.6336 KOps/s 40.1248 KOps/s $\color{#d91a1a}-3.72\%$
test_step_mdp_speed[False-False-True-False-True] 54.1210μs 23.9223μs 41.8020 KOps/s 43.7021 KOps/s $\color{#d91a1a}-4.35\%$
test_step_mdp_speed[False-False-True-False-False] 51.6400μs 15.5318μs 64.3842 KOps/s 67.5475 KOps/s $\color{#d91a1a}-4.68\%$
test_step_mdp_speed[False-False-False-True-True] 81.2420μs 39.1767μs 25.5254 KOps/s 26.0833 KOps/s $\color{#d91a1a}-2.14\%$
test_step_mdp_speed[False-False-False-True-False] 59.4810μs 27.1440μs 36.8405 KOps/s 38.1850 KOps/s $\color{#d91a1a}-3.52\%$
test_step_mdp_speed[False-False-False-False-True] 54.2710μs 25.4491μs 39.2941 KOps/s 40.6661 KOps/s $\color{#d91a1a}-3.37\%$
test_step_mdp_speed[False-False-False-False-False] 53.1390μs 17.1109μs 58.4422 KOps/s 60.9432 KOps/s $\color{#d91a1a}-4.10\%$
test_values[generalized_advantage_estimate-True-True] 10.9582ms 9.5091ms 105.1628 Ops/s 104.7705 Ops/s $\color{#35bf28}+0.37\%$
test_values[vec_generalized_advantage_estimate-True-True] 25.5025ms 23.9264ms 41.7949 Ops/s 41.3027 Ops/s $\color{#35bf28}+1.19\%$
test_values[td0_return_estimate-False-False] 0.2301ms 0.1758ms 5.6870 KOps/s 5.4374 KOps/s $\color{#35bf28}+4.59\%$
test_values[td1_return_estimate-False-False] 25.1874ms 23.6669ms 42.2531 Ops/s 41.8784 Ops/s $\color{#35bf28}+0.89\%$
test_values[vec_td1_return_estimate-False-False] 26.5057ms 24.0101ms 41.6492 Ops/s 41.3375 Ops/s $\color{#35bf28}+0.75\%$
test_values[td_lambda_return_estimate-True-False] 37.0329ms 34.1815ms 29.2556 Ops/s 28.9151 Ops/s $\color{#35bf28}+1.18\%$
test_values[vec_td_lambda_return_estimate-True-False] 25.6035ms 24.0624ms 41.5586 Ops/s 41.5259 Ops/s $\color{#35bf28}+0.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.4008ms 8.3846ms 119.2656 Ops/s 119.8799 Ops/s $\color{#d91a1a}-0.51\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3113ms 1.9093ms 523.7638 Ops/s 558.3742 Ops/s $\textbf{\color{#d91a1a}-6.20\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4698ms 0.3639ms 2.7483 KOps/s 2.7528 KOps/s $\color{#d91a1a}-0.16\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 53.4227ms 42.8158ms 23.3559 Ops/s 22.6767 Ops/s $\color{#35bf28}+3.00\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.1271ms 3.4214ms 292.2745 Ops/s 268.8986 Ops/s $\textbf{\color{#35bf28}+8.69\%}$
test_dqn_speed[False-None] 1.8829ms 1.4142ms 707.1313 Ops/s 713.0129 Ops/s $\color{#d91a1a}-0.82\%$
test_dqn_speed[False-backward] 1.9460ms 1.8939ms 528.0205 Ops/s 529.0442 Ops/s $\color{#d91a1a}-0.19\%$
test_dqn_speed[True-None] 0.7717ms 0.4817ms 2.0759 KOps/s 2.0846 KOps/s $\color{#d91a1a}-0.42\%$
test_dqn_speed[True-backward] 0.9512ms 0.9009ms 1.1100 KOps/s 1.0932 KOps/s $\color{#35bf28}+1.53\%$
test_dqn_speed[reduce-overhead-None] 0.5939ms 0.4831ms 2.0701 KOps/s 2.0619 KOps/s $\color{#35bf28}+0.40\%$
test_dqn_speed[reduce-overhead-backward] 0.9619ms 0.9070ms 1.1025 KOps/s 1.0948 KOps/s $\color{#35bf28}+0.71\%$
test_ddpg_speed[False-None] 3.6096ms 2.9015ms 344.6522 Ops/s 345.0588 Ops/s $\color{#d91a1a}-0.12\%$
test_ddpg_speed[False-backward] 4.1519ms 4.0295ms 248.1688 Ops/s 248.6335 Ops/s $\color{#d91a1a}-0.19\%$
test_ddpg_speed[True-None] 1.6158ms 1.2152ms 822.9244 Ops/s 814.1411 Ops/s $\color{#35bf28}+1.08\%$
test_ddpg_speed[True-backward] 2.1543ms 2.1085ms 474.2798 Ops/s 471.5156 Ops/s $\color{#35bf28}+0.59\%$
test_ddpg_speed[reduce-overhead-None] 1.4473ms 1.2111ms 825.7155 Ops/s 809.9717 Ops/s $\color{#35bf28}+1.94\%$
test_ddpg_speed[reduce-overhead-backward] 2.1790ms 2.1038ms 475.3387 Ops/s 462.8703 Ops/s $\color{#35bf28}+2.69\%$
test_sac_speed[False-None] 8.7728ms 8.0328ms 124.4895 Ops/s 124.2977 Ops/s $\color{#35bf28}+0.15\%$
test_sac_speed[False-backward] 14.4262ms 10.9283ms 91.5059 Ops/s 93.5266 Ops/s $\color{#d91a1a}-2.16\%$
test_sac_speed[True-None] 2.3095ms 2.0784ms 481.1302 Ops/s 480.5361 Ops/s $\color{#35bf28}+0.12\%$
test_sac_speed[True-backward] 4.0083ms 3.7602ms 265.9448 Ops/s 267.3004 Ops/s $\color{#d91a1a}-0.51\%$
test_sac_speed[reduce-overhead-None] 2.3327ms 2.0743ms 482.0809 Ops/s 477.2514 Ops/s $\color{#35bf28}+1.01\%$
test_sac_speed[reduce-overhead-backward] 3.8604ms 3.7721ms 265.1012 Ops/s 264.6680 Ops/s $\color{#35bf28}+0.16\%$
test_redq_speed[False-None] 14.5614ms 12.8976ms 77.5337 Ops/s 78.4664 Ops/s $\color{#d91a1a}-1.19\%$
test_redq_speed[False-backward] 24.9071ms 22.0116ms 45.4305 Ops/s 45.0303 Ops/s $\color{#35bf28}+0.89\%$
test_redq_speed[True-None] 5.5214ms 4.7595ms 210.1066 Ops/s 210.0555 Ops/s $\color{#35bf28}+0.02\%$
test_redq_speed[True-backward] 13.1425ms 12.0536ms 82.9626 Ops/s 77.1248 Ops/s $\textbf{\color{#35bf28}+7.57\%}$
test_redq_speed[reduce-overhead-None] 5.4870ms 4.7560ms 210.2613 Ops/s 208.6745 Ops/s $\color{#35bf28}+0.76\%$
test_redq_speed[reduce-overhead-backward] 13.7057ms 12.1765ms 82.1253 Ops/s 73.7234 Ops/s $\textbf{\color{#35bf28}+11.40\%}$
test_redq_deprec_speed[False-None] 14.5292ms 12.6658ms 78.9530 Ops/s 71.8220 Ops/s $\textbf{\color{#35bf28}+9.93\%}$
test_redq_deprec_speed[False-backward] 19.7787ms 18.3162ms 54.5964 Ops/s 52.6270 Ops/s $\color{#35bf28}+3.74\%$
test_redq_deprec_speed[True-None] 4.6630ms 3.8216ms 261.6724 Ops/s 262.2634 Ops/s $\color{#d91a1a}-0.23\%$
test_redq_deprec_speed[True-backward] 8.2214ms 8.1296ms 123.0080 Ops/s 123.1432 Ops/s $\color{#d91a1a}-0.11\%$
test_redq_deprec_speed[reduce-overhead-None] 4.5662ms 3.8056ms 262.7683 Ops/s 263.0501 Ops/s $\color{#d91a1a}-0.11\%$
test_redq_deprec_speed[reduce-overhead-backward] 10.2221ms 8.2220ms 121.6246 Ops/s 122.6227 Ops/s $\color{#d91a1a}-0.81\%$
test_td3_speed[False-None] 8.3810ms 8.0765ms 123.8164 Ops/s 125.5385 Ops/s $\color{#d91a1a}-1.37\%$
test_td3_speed[False-backward] 10.9060ms 10.4812ms 95.4091 Ops/s 96.5826 Ops/s $\color{#d91a1a}-1.22\%$
test_td3_speed[True-None] 1.8980ms 1.7766ms 562.8624 Ops/s 568.6106 Ops/s $\color{#d91a1a}-1.01\%$
test_td3_speed[True-backward] 3.4373ms 3.3690ms 296.8237 Ops/s 300.1777 Ops/s $\color{#d91a1a}-1.12\%$
test_td3_speed[reduce-overhead-None] 1.8954ms 1.7776ms 562.5688 Ops/s 566.9270 Ops/s $\color{#d91a1a}-0.77\%$
test_td3_speed[reduce-overhead-backward] 4.4778ms 3.4088ms 293.3597 Ops/s 298.7833 Ops/s $\color{#d91a1a}-1.82\%$
test_cql_speed[False-None] 41.9043ms 36.5963ms 27.3252 Ops/s 27.5302 Ops/s $\color{#d91a1a}-0.74\%$
test_cql_speed[False-backward] 49.1023ms 46.4707ms 21.5189 Ops/s 21.7732 Ops/s $\color{#d91a1a}-1.17\%$
test_cql_speed[True-None] 17.3893ms 15.7943ms 63.3139 Ops/s 60.3634 Ops/s $\color{#35bf28}+4.89\%$
test_cql_speed[True-backward] 23.7533ms 22.3795ms 44.6838 Ops/s 42.9465 Ops/s $\color{#35bf28}+4.05\%$
test_cql_speed[reduce-overhead-None] 18.1875ms 15.9259ms 62.7908 Ops/s 62.7551 Ops/s $\color{#35bf28}+0.06\%$
test_cql_speed[reduce-overhead-backward] 24.1280ms 22.3270ms 44.7889 Ops/s 42.7051 Ops/s $\color{#35bf28}+4.88\%$
test_a2c_speed[False-None] 8.3817ms 7.1491ms 139.8781 Ops/s 139.9689 Ops/s $\color{#d91a1a}-0.06\%$
test_a2c_speed[False-backward] 16.3389ms 14.1346ms 70.7482 Ops/s 70.4970 Ops/s $\color{#35bf28}+0.36\%$
test_a2c_speed[True-None] 4.1622ms 3.7070ms 269.7605 Ops/s 268.6601 Ops/s $\color{#35bf28}+0.41\%$
test_a2c_speed[True-backward] 10.8028ms 10.1494ms 98.5277 Ops/s 98.7011 Ops/s $\color{#d91a1a}-0.18\%$
test_a2c_speed[reduce-overhead-None] 4.5195ms 3.7055ms 269.8666 Ops/s 270.8206 Ops/s $\color{#d91a1a}-0.35\%$
test_a2c_speed[reduce-overhead-backward] 11.3861ms 10.0975ms 99.0345 Ops/s 99.0133 Ops/s $\color{#35bf28}+0.02\%$
test_ppo_speed[False-None] 9.1344ms 7.4888ms 133.5331 Ops/s 134.7351 Ops/s $\color{#d91a1a}-0.89\%$
test_ppo_speed[False-backward] 16.2283ms 14.7113ms 67.9751 Ops/s 69.1516 Ops/s $\color{#d91a1a}-1.70\%$
test_ppo_speed[True-None] 4.5184ms 4.0615ms 246.2153 Ops/s 245.9414 Ops/s $\color{#35bf28}+0.11\%$
test_ppo_speed[True-backward] 10.2840ms 9.9106ms 100.9017 Ops/s 97.4254 Ops/s $\color{#35bf28}+3.57\%$
test_ppo_speed[reduce-overhead-None] 5.2679ms 4.0596ms 246.3289 Ops/s 246.2181 Ops/s $\color{#35bf28}+0.04\%$
test_ppo_speed[reduce-overhead-backward] 10.2251ms 9.8991ms 101.0189 Ops/s 100.0415 Ops/s $\color{#35bf28}+0.98\%$
test_reinforce_speed[False-None] 7.9064ms 6.5379ms 152.9542 Ops/s 152.5234 Ops/s $\color{#35bf28}+0.28\%$
test_reinforce_speed[False-backward] 11.0746ms 9.7950ms 102.0931 Ops/s 102.7644 Ops/s $\color{#d91a1a}-0.65\%$
test_reinforce_speed[True-None] 3.7331ms 3.0268ms 330.3869 Ops/s 330.4177 Ops/s $-0.01\%$
test_reinforce_speed[True-backward] 9.5228ms 8.9114ms 112.2162 Ops/s 111.6424 Ops/s $\color{#35bf28}+0.51\%$
test_reinforce_speed[reduce-overhead-None] 4.5980ms 3.0606ms 326.7376 Ops/s 329.3474 Ops/s $\color{#d91a1a}-0.79\%$
test_reinforce_speed[reduce-overhead-backward] 10.5585ms 8.9769ms 111.3967 Ops/s 111.8619 Ops/s $\color{#d91a1a}-0.42\%$
test_iql_speed[False-None] 33.9279ms 32.2847ms 30.9744 Ops/s 30.7602 Ops/s $\color{#35bf28}+0.70\%$
test_iql_speed[False-backward] 48.1998ms 45.3998ms 22.0266 Ops/s 21.8241 Ops/s $\color{#35bf28}+0.93\%$
test_iql_speed[True-None] 12.8434ms 11.0087ms 90.8373 Ops/s 89.6088 Ops/s $\color{#35bf28}+1.37\%$
test_iql_speed[True-backward] 22.7218ms 21.6616ms 46.1647 Ops/s 45.4902 Ops/s $\color{#35bf28}+1.48\%$
test_iql_speed[reduce-overhead-None] 12.3270ms 11.0581ms 90.4312 Ops/s 90.1576 Ops/s $\color{#35bf28}+0.30\%$
test_iql_speed[reduce-overhead-backward] 23.2130ms 21.7492ms 45.9786 Ops/s 45.8666 Ops/s $\color{#35bf28}+0.24\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8597ms 4.7497ms 210.5388 Ops/s 210.7649 Ops/s $\color{#d91a1a}-0.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7725ms 0.5117ms 1.9543 KOps/s 1.9717 KOps/s $\color{#d91a1a}-0.88\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8851ms 0.4931ms 2.0282 KOps/s 2.0656 KOps/s $\color{#d91a1a}-1.81\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.1707ms 4.5294ms 220.7797 Ops/s 223.0907 Ops/s $\color{#d91a1a}-1.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.9656ms 0.5070ms 1.9725 KOps/s 1.9851 KOps/s $\color{#d91a1a}-0.63\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6840ms 0.4801ms 2.0828 KOps/s 2.0955 KOps/s $\color{#d91a1a}-0.61\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.7788ms 1.6590ms 602.7642 Ops/s 613.7574 Ops/s $\color{#d91a1a}-1.79\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.3671ms 1.5759ms 634.5411 Ops/s 648.1877 Ops/s $\color{#d91a1a}-2.11\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.5467ms 4.7241ms 211.6790 Ops/s 214.4924 Ops/s $\color{#d91a1a}-1.31\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.4682ms 0.6538ms 1.5295 KOps/s 1.5627 KOps/s $\color{#d91a1a}-2.13\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0063ms 0.6305ms 1.5861 KOps/s 1.6145 KOps/s $\color{#d91a1a}-1.76\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 4.7374ms 4.5650ms 219.0563 Ops/s 221.1462 Ops/s $\color{#d91a1a}-0.95\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1501ms 0.5226ms 1.9135 KOps/s 1.9863 KOps/s $\color{#d91a1a}-3.67\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7017ms 0.4872ms 2.0525 KOps/s 2.0625 KOps/s $\color{#d91a1a}-0.48\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7076ms 4.5704ms 218.7978 Ops/s 225.6320 Ops/s $\color{#d91a1a}-3.03\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9339ms 0.5144ms 1.9439 KOps/s 1.9740 KOps/s $\color{#d91a1a}-1.53\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7524ms 0.4834ms 2.0689 KOps/s 2.1123 KOps/s $\color{#d91a1a}-2.06\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.2784ms 4.6969ms 212.9068 Ops/s 216.4427 Ops/s $\color{#d91a1a}-1.63\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2403ms 0.6598ms 1.5155 KOps/s 1.5625 KOps/s $\color{#d91a1a}-3.00\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8399ms 0.6277ms 1.5930 KOps/s 1.6211 KOps/s $\color{#d91a1a}-1.73\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.4864ms 4.2263ms 236.6144 Ops/s 248.3039 Ops/s $\color{#d91a1a}-4.71\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 12.5645ms 2.7015ms 370.1663 Ops/s 448.6240 Ops/s $\textbf{\color{#d91a1a}-17.49\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.4620ms 1.3321ms 750.7022 Ops/s 757.8383 Ops/s $\color{#d91a1a}-0.94\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 5.5017ms 4.2514ms 235.2172 Ops/s 35.2488 Ops/s $\textbf{\color{#35bf28}+567.31\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.5243ms 2.3337ms 428.4983 Ops/s 430.1436 Ops/s $\color{#d91a1a}-0.38\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 3.8977ms 1.2823ms 779.8311 Ops/s 732.2506 Ops/s $\textbf{\color{#35bf28}+6.50\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4197s 12.6896ms 78.8049 Ops/s 226.1794 Ops/s $\textbf{\color{#d91a1a}-65.16\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.9980ms 2.4467ms 408.7218 Ops/s 405.5216 Ops/s $\color{#35bf28}+0.79\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.6825ms 1.5020ms 665.7790 Ops/s 634.2642 Ops/s $\color{#35bf28}+4.97\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 11.9206ms 11.5550ms 86.5425 Ops/s 86.4556 Ops/s $\color{#35bf28}+0.10\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.6675ms 14.5773ms 68.5998 Ops/s 70.0134 Ops/s $\color{#d91a1a}-2.02\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 22.3713ms 20.3057ms 49.2473 Ops/s 49.0254 Ops/s $\color{#35bf28}+0.45\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 15.5402ms 14.6388ms 68.3115 Ops/s 69.7859 Ops/s $\color{#d91a1a}-2.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 22.7312ms 20.3505ms 49.1388 Ops/s 49.3997 Ops/s $\color{#d91a1a}-0.53\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 17.2737ms 15.8485ms 63.0975 Ops/s 63.6753 Ops/s $\color{#d91a1a}-0.91\%$

Copy link

github-actions bot commented Feb 3, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8312s 0.7461s 1.3404 Ops/s 1.3709 Ops/s $\color{#d91a1a}-2.23\%$
test_transformed 1.4287s 1.3435s 0.7443 Ops/s 0.7356 Ops/s $\color{#35bf28}+1.18\%$
test_serial 2.1538s 2.1471s 0.4657 Ops/s 0.4510 Ops/s $\color{#35bf28}+3.27\%$
test_parallel 1.8685s 1.8413s 0.5431 Ops/s 0.5464 Ops/s $\color{#d91a1a}-0.61\%$
test_step_mdp_speed[True-True-True-True-True] 0.1413ms 38.7716μs 25.7920 KOps/s 25.3728 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-True-True-True-False] 0.1555ms 23.2614μs 42.9897 KOps/s 41.9901 KOps/s $\color{#35bf28}+2.38\%$
test_step_mdp_speed[True-True-True-False-True] 52.0410μs 21.8998μs 45.6626 KOps/s 43.5776 KOps/s $\color{#35bf28}+4.78\%$
test_step_mdp_speed[True-True-True-False-False] 46.5900μs 12.9532μs 77.2009 KOps/s 74.5607 KOps/s $\color{#35bf28}+3.54\%$
test_step_mdp_speed[True-True-False-True-True] 0.1036ms 42.4278μs 23.5695 KOps/s 23.1856 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[True-True-False-True-False] 0.1462ms 25.5990μs 39.0640 KOps/s 37.9707 KOps/s $\color{#35bf28}+2.88\%$
test_step_mdp_speed[True-True-False-False-True] 61.1210μs 25.1617μs 39.7430 KOps/s 39.4595 KOps/s $\color{#35bf28}+0.72\%$
test_step_mdp_speed[True-True-False-False-False] 45.4710μs 15.3684μs 65.0684 KOps/s 63.1880 KOps/s $\color{#35bf28}+2.98\%$
test_step_mdp_speed[True-False-True-True-True] 80.6210μs 44.8468μs 22.2982 KOps/s 21.7861 KOps/s $\color{#35bf28}+2.35\%$
test_step_mdp_speed[True-False-True-True-False] 89.3820μs 27.0731μs 36.9370 KOps/s 35.3455 KOps/s $\color{#35bf28}+4.50\%$
test_step_mdp_speed[True-False-True-False-True] 62.1110μs 24.1943μs 41.3321 KOps/s 39.1816 KOps/s $\textbf{\color{#35bf28}+5.49\%}$
test_step_mdp_speed[True-False-True-False-False] 0.1924ms 15.4095μs 64.8952 KOps/s 63.8427 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-False-False-True-True] 84.3210μs 46.2315μs 21.6303 KOps/s 20.8710 KOps/s $\color{#35bf28}+3.64\%$
test_step_mdp_speed[True-False-False-True-False] 63.6910μs 29.7518μs 33.6114 KOps/s 32.3830 KOps/s $\color{#35bf28}+3.79\%$
test_step_mdp_speed[True-False-False-False-True] 88.0010μs 26.4771μs 37.7685 KOps/s 36.6697 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[True-False-False-False-False] 49.5810μs 17.4224μs 57.3974 KOps/s 55.0377 KOps/s $\color{#35bf28}+4.29\%$
test_step_mdp_speed[False-True-True-True-True] 76.9010μs 45.1573μs 22.1448 KOps/s 21.6721 KOps/s $\color{#35bf28}+2.18\%$
test_step_mdp_speed[False-True-True-True-False] 56.0210μs 27.7292μs 36.0630 KOps/s 34.8903 KOps/s $\color{#35bf28}+3.36\%$
test_step_mdp_speed[False-True-True-False-True] 2.7801ms 28.6926μs 34.8522 KOps/s 33.8696 KOps/s $\color{#35bf28}+2.90\%$
test_step_mdp_speed[False-True-True-False-False] 75.7410μs 17.2317μs 58.0327 KOps/s 57.7599 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[False-True-False-True-True] 74.7410μs 46.5802μs 21.4683 KOps/s 20.9281 KOps/s $\color{#35bf28}+2.58\%$
test_step_mdp_speed[False-True-False-True-False] 60.2610μs 30.0440μs 33.2845 KOps/s 32.1947 KOps/s $\color{#35bf28}+3.39\%$
test_step_mdp_speed[False-True-False-False-True] 61.6610μs 30.9984μs 32.2597 KOps/s 31.1525 KOps/s $\color{#35bf28}+3.55\%$
test_step_mdp_speed[False-True-False-False-False] 49.6910μs 19.3645μs 51.6410 KOps/s 50.6583 KOps/s $\color{#35bf28}+1.94\%$
test_step_mdp_speed[False-False-True-True-True] 81.1710μs 49.3866μs 20.2484 KOps/s 20.0398 KOps/s $\color{#35bf28}+1.04\%$
test_step_mdp_speed[False-False-True-True-False] 61.7110μs 32.6887μs 30.5917 KOps/s 30.2685 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-False-True-False-True] 65.2020μs 30.6592μs 32.6167 KOps/s 32.4192 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[False-False-True-False-False] 52.3710μs 19.3459μs 51.6906 KOps/s 50.3470 KOps/s $\color{#35bf28}+2.67\%$
test_step_mdp_speed[False-False-False-True-True] 83.3420μs 50.9761μs 19.6170 KOps/s 19.1374 KOps/s $\color{#35bf28}+2.51\%$
test_step_mdp_speed[False-False-False-True-False] 75.9910μs 34.8765μs 28.6726 KOps/s 28.1935 KOps/s $\color{#35bf28}+1.70\%$
test_step_mdp_speed[False-False-False-False-True] 65.8410μs 32.5114μs 30.7585 KOps/s 30.6087 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[False-False-False-False-False] 52.6110μs 21.5629μs 46.3759 KOps/s 45.3438 KOps/s $\color{#35bf28}+2.28\%$
test_values[generalized_advantage_estimate-True-True] 25.9845ms 25.4158ms 39.3456 Ops/s 40.0708 Ops/s $\color{#d91a1a}-1.81\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1233s 3.3653ms 297.1483 Ops/s 340.1236 Ops/s $\textbf{\color{#d91a1a}-12.64\%}$
test_values[td0_return_estimate-False-False] 0.1182ms 80.0496μs 12.4923 KOps/s 12.5098 KOps/s $\color{#d91a1a}-0.14\%$
test_values[td1_return_estimate-False-False] 60.8120ms 57.5656ms 17.3715 Ops/s 17.9746 Ops/s $\color{#d91a1a}-3.36\%$
test_values[vec_td1_return_estimate-False-False] 1.3179ms 1.0870ms 919.9619 Ops/s 920.6143 Ops/s $\color{#d91a1a}-0.07\%$
test_values[td_lambda_return_estimate-True-False] 92.7085ms 89.3672ms 11.1898 Ops/s 11.3949 Ops/s $\color{#d91a1a}-1.80\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4373ms 1.0846ms 922.0109 Ops/s 920.3126 Ops/s $\color{#35bf28}+0.18\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.5227ms 25.1772ms 39.7185 Ops/s 40.2016 Ops/s $\color{#d91a1a}-1.20\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0479ms 0.7570ms 1.3210 KOps/s 1.3294 KOps/s $\color{#d91a1a}-0.63\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7644ms 0.6735ms 1.4847 KOps/s 1.4950 KOps/s $\color{#d91a1a}-0.69\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6429ms 1.4895ms 671.3485 Ops/s 671.5130 Ops/s $\color{#d91a1a}-0.02\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7278ms 0.6845ms 1.4608 KOps/s 1.4622 KOps/s $\color{#d91a1a}-0.09\%$
test_dqn_speed[False-None] 1.6510ms 1.5235ms 656.3831 Ops/s 643.6192 Ops/s $\color{#35bf28}+1.98\%$
test_dqn_speed[False-backward] 2.2749ms 2.1486ms 465.4149 Ops/s 461.5523 Ops/s $\color{#35bf28}+0.84\%$
test_dqn_speed[True-None] 0.7126ms 0.5577ms 1.7932 KOps/s 1.7737 KOps/s $\color{#35bf28}+1.10\%$
test_dqn_speed[True-backward] 1.2811ms 1.2224ms 818.0833 Ops/s 879.6875 Ops/s $\textbf{\color{#d91a1a}-7.00\%}$
test_dqn_speed[reduce-overhead-None] 0.6287ms 0.5779ms 1.7303 KOps/s 1.7303 KOps/s $+0.00\%$
test_dqn_speed[reduce-overhead-backward] 1.1407ms 1.0743ms 930.8113 Ops/s 1.0258 KOps/s $\textbf{\color{#d91a1a}-9.26\%}$
test_ddpg_speed[False-None] 3.2411ms 2.8989ms 344.9600 Ops/s 340.8697 Ops/s $\color{#35bf28}+1.20\%$
test_ddpg_speed[False-backward] 4.8243ms 4.3165ms 231.6718 Ops/s 235.6586 Ops/s $\color{#d91a1a}-1.69\%$
test_ddpg_speed[True-None] 1.5082ms 1.3370ms 747.9249 Ops/s 740.1193 Ops/s $\color{#35bf28}+1.05\%$
test_ddpg_speed[True-backward] 2.7033ms 2.5640ms 390.0170 Ops/s 406.7406 Ops/s $\color{#d91a1a}-4.11\%$
test_ddpg_speed[reduce-overhead-None] 1.5378ms 1.3522ms 739.5549 Ops/s 729.0035 Ops/s $\color{#35bf28}+1.45\%$
test_ddpg_speed[reduce-overhead-backward] 2.1919ms 2.0330ms 491.8761 Ops/s 523.0017 Ops/s $\textbf{\color{#d91a1a}-5.95\%}$
test_sac_speed[False-None] 8.5541ms 8.1136ms 123.2496 Ops/s 122.5031 Ops/s $\color{#35bf28}+0.61\%$
test_sac_speed[False-backward] 11.7224ms 11.2602ms 88.8082 Ops/s 90.0474 Ops/s $\color{#d91a1a}-1.38\%$
test_sac_speed[True-None] 1.8939ms 1.8341ms 545.2293 Ops/s 533.3529 Ops/s $\color{#35bf28}+2.23\%$
test_sac_speed[True-backward] 4.1719ms 3.7271ms 268.3029 Ops/s 264.3563 Ops/s $\color{#35bf28}+1.49\%$
test_sac_speed[reduce-overhead-None] 21.7402ms 12.1809ms 82.0954 Ops/s 83.4658 Ops/s $\color{#d91a1a}-1.64\%$
test_sac_speed[reduce-overhead-backward] 1.9147ms 1.7819ms 561.2067 Ops/s 547.5443 Ops/s $\color{#35bf28}+2.50\%$
test_redq_speed[False-None] 8.0093ms 7.5761ms 131.9937 Ops/s 129.8334 Ops/s $\color{#35bf28}+1.66\%$
test_redq_speed[False-backward] 14.0622ms 11.8221ms 84.5870 Ops/s 84.0540 Ops/s $\color{#35bf28}+0.63\%$
test_redq_speed[True-None] 2.4691ms 2.2963ms 435.4913 Ops/s 426.4934 Ops/s $\color{#35bf28}+2.11\%$
test_redq_speed[True-backward] 4.6410ms 4.1805ms 239.2053 Ops/s 234.8043 Ops/s $\color{#35bf28}+1.87\%$
test_redq_speed[reduce-overhead-None] 2.5084ms 2.3061ms 433.6372 Ops/s 422.9965 Ops/s $\color{#35bf28}+2.52\%$
test_redq_speed[reduce-overhead-backward] 4.3786ms 4.1716ms 239.7186 Ops/s 235.2199 Ops/s $\color{#35bf28}+1.91\%$
test_redq_deprec_speed[False-None] 9.4643ms 9.1310ms 109.5172 Ops/s 107.4905 Ops/s $\color{#35bf28}+1.89\%$
test_redq_deprec_speed[False-backward] 12.7833ms 12.3587ms 80.9149 Ops/s 80.1035 Ops/s $\color{#35bf28}+1.01\%$
test_redq_deprec_speed[True-None] 2.8641ms 2.6435ms 378.2920 Ops/s 373.5998 Ops/s $\color{#35bf28}+1.26\%$
test_redq_deprec_speed[True-backward] 4.8974ms 4.4710ms 223.6653 Ops/s 219.2462 Ops/s $\color{#35bf28}+2.02\%$
test_redq_deprec_speed[reduce-overhead-None] 2.7878ms 2.6263ms 380.7692 Ops/s 374.0906 Ops/s $\color{#35bf28}+1.79\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.8330ms 4.4807ms 223.1818 Ops/s 220.1277 Ops/s $\color{#35bf28}+1.39\%$
test_td3_speed[False-None] 8.3309ms 8.0688ms 123.9344 Ops/s 122.4132 Ops/s $\color{#35bf28}+1.24\%$
test_td3_speed[False-backward] 11.4353ms 10.6672ms 93.7451 Ops/s 93.9141 Ops/s $\color{#d91a1a}-0.18\%$
test_td3_speed[True-None] 1.6573ms 1.6336ms 612.1394 Ops/s 587.3199 Ops/s $\color{#35bf28}+4.23\%$
test_td3_speed[True-backward] 3.3688ms 3.3004ms 302.9978 Ops/s 289.5882 Ops/s $\color{#35bf28}+4.63\%$
test_td3_speed[reduce-overhead-None] 55.1400ms 26.6498ms 37.5238 Ops/s 36.2530 Ops/s $\color{#35bf28}+3.51\%$
test_td3_speed[reduce-overhead-backward] 1.5321ms 1.4960ms 668.4661 Ops/s 649.5188 Ops/s $\color{#35bf28}+2.92\%$
test_cql_speed[False-None] 17.8690ms 17.0050ms 58.8061 Ops/s 58.7341 Ops/s $\color{#35bf28}+0.12\%$
test_cql_speed[False-backward] 23.3656ms 22.4595ms 44.5245 Ops/s 44.3152 Ops/s $\color{#35bf28}+0.47\%$
test_cql_speed[True-None] 3.3397ms 3.2511ms 307.5877 Ops/s 301.0369 Ops/s $\color{#35bf28}+2.18\%$
test_cql_speed[True-backward] 6.3631ms 5.7197ms 174.8352 Ops/s 175.0615 Ops/s $\color{#d91a1a}-0.13\%$
test_cql_speed[reduce-overhead-None] 20.8947ms 13.1525ms 76.0312 Ops/s 57.4557 Ops/s $\textbf{\color{#35bf28}+32.33\%}$
test_cql_speed[reduce-overhead-backward] 2.1394ms 1.9994ms 500.1592 Ops/s 492.2284 Ops/s $\color{#35bf28}+1.61\%$
test_a2c_speed[False-None] 3.4242ms 3.2245ms 310.1272 Ops/s 305.9249 Ops/s $\color{#35bf28}+1.37\%$
test_a2c_speed[False-backward] 6.9774ms 6.3641ms 157.1324 Ops/s 155.6942 Ops/s $\color{#35bf28}+0.92\%$
test_a2c_speed[True-None] 1.5322ms 1.3474ms 742.1934 Ops/s 735.6257 Ops/s $\color{#35bf28}+0.89\%$
test_a2c_speed[True-backward] 3.1052ms 2.8730ms 348.0694 Ops/s 320.2346 Ops/s $\textbf{\color{#35bf28}+8.69\%}$
test_a2c_speed[reduce-overhead-None] 15.8605ms 9.0666ms 110.2945 Ops/s 110.3089 Ops/s $\color{#d91a1a}-0.01\%$
test_a2c_speed[reduce-overhead-backward] 1.8586ms 1.4698ms 680.3754 Ops/s 611.6612 Ops/s $\textbf{\color{#35bf28}+11.23\%}$
test_ppo_speed[False-None] 4.0811ms 3.7129ms 269.3296 Ops/s 264.4126 Ops/s $\color{#35bf28}+1.86\%$
test_ppo_speed[False-backward] 7.2737ms 6.8765ms 145.4221 Ops/s 139.7251 Ops/s $\color{#35bf28}+4.08\%$
test_ppo_speed[True-None] 1.8148ms 1.4066ms 710.9185 Ops/s 697.2220 Ops/s $\color{#35bf28}+1.96\%$
test_ppo_speed[True-backward] 3.1144ms 3.0512ms 327.7411 Ops/s 304.0436 Ops/s $\textbf{\color{#35bf28}+7.79\%}$
test_ppo_speed[reduce-overhead-None] 1.3692ms 0.9761ms 1.0245 KOps/s 1.0199 KOps/s $\color{#35bf28}+0.45\%$
test_ppo_speed[reduce-overhead-backward] 1.5531ms 1.4067ms 710.8827 Ops/s 616.5578 Ops/s $\textbf{\color{#35bf28}+15.30\%}$
test_reinforce_speed[False-None] 2.7022ms 2.2969ms 435.3613 Ops/s 427.0719 Ops/s $\color{#35bf28}+1.94\%$
test_reinforce_speed[False-backward] 3.9589ms 3.3125ms 301.8909 Ops/s 288.1346 Ops/s $\color{#35bf28}+4.77\%$
test_reinforce_speed[True-None] 1.6901ms 1.2942ms 772.6949 Ops/s 757.5530 Ops/s $\color{#35bf28}+2.00\%$
test_reinforce_speed[True-backward] 3.0054ms 2.9179ms 342.7125 Ops/s 324.0763 Ops/s $\textbf{\color{#35bf28}+5.75\%}$
test_reinforce_speed[reduce-overhead-None] 18.4122ms 10.1628ms 98.3982 Ops/s 99.9229 Ops/s $\color{#d91a1a}-1.53\%$
test_reinforce_speed[reduce-overhead-backward] 1.6072ms 1.5656ms 638.7328 Ops/s 599.2375 Ops/s $\textbf{\color{#35bf28}+6.59\%}$
test_iql_speed[False-None] 9.8203ms 9.2635ms 107.9503 Ops/s 104.8946 Ops/s $\color{#35bf28}+2.91\%$
test_iql_speed[False-backward] 13.3675ms 13.0265ms 76.7663 Ops/s 74.4314 Ops/s $\color{#35bf28}+3.14\%$
test_iql_speed[True-None] 2.4215ms 2.2298ms 448.4744 Ops/s 431.0275 Ops/s $\color{#35bf28}+4.05\%$
test_iql_speed[True-backward] 5.3875ms 4.9072ms 203.7806 Ops/s 199.1668 Ops/s $\color{#35bf28}+2.32\%$
test_iql_speed[reduce-overhead-None] 18.9046ms 11.2165ms 89.1545 Ops/s 88.7629 Ops/s $\color{#35bf28}+0.44\%$
test_iql_speed[reduce-overhead-backward] 2.2399ms 2.0626ms 484.8269 Ops/s 461.7544 Ops/s $\color{#35bf28}+5.00\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0883ms 6.3496ms 157.4904 Ops/s 154.0120 Ops/s $\color{#35bf28}+2.26\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6449ms 0.3755ms 2.6633 KOps/s 3.1630 KOps/s $\textbf{\color{#d91a1a}-15.80\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7552ms 0.3575ms 2.7974 KOps/s 3.2896 KOps/s $\textbf{\color{#d91a1a}-14.96\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3995ms 6.0934ms 164.1121 Ops/s 163.4174 Ops/s $\color{#35bf28}+0.43\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0343ms 0.3159ms 3.1653 KOps/s 3.3845 KOps/s $\textbf{\color{#d91a1a}-6.47\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6896ms 0.2829ms 3.5352 KOps/s 3.3488 KOps/s $\textbf{\color{#35bf28}+5.57\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5538ms 1.3222ms 756.3240 Ops/s 715.8734 Ops/s $\textbf{\color{#35bf28}+5.65\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5375ms 1.2746ms 784.5492 Ops/s 775.8732 Ops/s $\color{#35bf28}+1.12\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4745ms 6.2850ms 159.1086 Ops/s 157.0040 Ops/s $\color{#35bf28}+1.34\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9400ms 0.4148ms 2.4109 KOps/s 2.4059 KOps/s $\color{#35bf28}+0.21\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6400ms 0.3942ms 2.5370 KOps/s 2.4882 KOps/s $\color{#35bf28}+1.96\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.3280ms 6.1211ms 163.3706 Ops/s 163.5383 Ops/s $\color{#d91a1a}-0.10\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8919ms 0.2781ms 3.5963 KOps/s 2.9564 KOps/s $\textbf{\color{#35bf28}+21.65\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5128ms 0.2791ms 3.5827 KOps/s 3.1898 KOps/s $\textbf{\color{#35bf28}+12.32\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 10.0926ms 6.1499ms 162.6037 Ops/s 163.3647 Ops/s $\color{#d91a1a}-0.47\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7086ms 0.3136ms 3.1890 KOps/s 2.7459 KOps/s $\textbf{\color{#35bf28}+16.14\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5356ms 0.2859ms 3.4980 KOps/s 2.8350 KOps/s $\textbf{\color{#35bf28}+23.39\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4839ms 6.2329ms 160.4382 Ops/s 157.0478 Ops/s $\color{#35bf28}+2.16\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1771ms 0.4133ms 2.4197 KOps/s 2.2222 KOps/s $\textbf{\color{#35bf28}+8.89\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6597ms 0.4402ms 2.2718 KOps/s 2.2898 KOps/s $\color{#d91a1a}-0.79\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1791ms 5.5064ms 181.6070 Ops/s 177.4532 Ops/s $\color{#35bf28}+2.34\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 6.5606ms 2.1166ms 472.4570 Ops/s 425.2128 Ops/s $\textbf{\color{#35bf28}+11.11\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 9.2210ms 1.3141ms 761.0036 Ops/s 827.8329 Ops/s $\textbf{\color{#d91a1a}-8.07\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 8.4028ms 5.6847ms 175.9102 Ops/s 181.3557 Ops/s $\color{#d91a1a}-3.00\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.7560ms 2.1115ms 473.6027 Ops/s 450.1619 Ops/s $\textbf{\color{#35bf28}+5.21\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.2718ms 1.3325ms 750.4821 Ops/s 771.0656 Ops/s $\color{#d91a1a}-2.67\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5022s 15.7751ms 63.3911 Ops/s 31.2406 Ops/s $\textbf{\color{#35bf28}+102.91\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.0089ms 2.3159ms 431.8058 Ops/s 463.9687 Ops/s $\textbf{\color{#d91a1a}-6.93\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.6156ms 1.4308ms 698.8891 Ops/s 845.9581 Ops/s $\textbf{\color{#d91a1a}-17.38\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.7264ms 13.0588ms 76.5769 Ops/s 72.8810 Ops/s $\textbf{\color{#35bf28}+5.07\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.3274ms 16.7766ms 59.6068 Ops/s 58.2880 Ops/s $\color{#35bf28}+2.26\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.1898ms 17.7700ms 56.2747 Ops/s 54.5327 Ops/s $\color{#35bf28}+3.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.4674ms 16.9408ms 59.0290 Ops/s 57.6204 Ops/s $\color{#35bf28}+2.44\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.1735ms 17.7412ms 56.3660 Ops/s 55.0828 Ops/s $\color{#35bf28}+2.33\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.3499ms 18.8274ms 53.1140 Ops/s 53.1202 Ops/s $\color{#d91a1a}-0.01\%$

[ghstack-poisoned]
@vmoens vmoens added bc breaking backward compatibility breaking change Deprecation labels Feb 4, 2025
@vmoens vmoens merged commit 6f16f50 into gh/vmoens/93/base Feb 4, 2025
36 of 51 checks passed
vmoens added a commit that referenced this pull request Feb 4, 2025
ghstack-source-id: eeae1bf0611a5d293f533767eee7b9700e720cc8
Pull Request resolved: #2750
@vmoens vmoens deleted the gh/vmoens/93/head branch February 4, 2025 08:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bc breaking backward compatibility breaking change CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Deprecation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants