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

Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Jul 24, 2025

No description provided.

@pytorch-bot
Copy link

pytorch-bot bot commented Jul 24, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3092

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

❌ 1 New Failure, 2 Cancelled Jobs, 5 Unrelated Failures

As of commit 214a215 with merge base 2345de0 (image):

NEW FAILURE - The following job has failed:

CANCELLED JOBS - The following jobs were cancelled. Please retry:

FLAKY - The following job failed but was likely due to flakiness present on trunk:

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

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

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

@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 24, 2025
@vmoens vmoens added CI Has to do with CI setup (e.g. wheels & builds, tests...) Benchmarks rl/benchmark changes labels Jul 24, 2025
@github-actions
Copy link

github-actions bot commented Jul 24, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5372s 0.5358s 1.8662 Ops/s 1.8046 Ops/s $\color{#35bf28}+3.41\%$
test_transformed 1.1014s 1.0983s 0.9105 Ops/s 0.8993 Ops/s $\color{#35bf28}+1.25\%$
test_serial 1.6472s 1.6440s 0.6083 Ops/s 0.6026 Ops/s $\color{#35bf28}+0.95\%$
test_parallel 1.1111s 1.0558s 0.9471 Ops/s 0.9387 Ops/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[True-True-True-True-True] 0.2659ms 44.0364μs 22.7085 KOps/s 22.7394 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[True-True-True-True-False] 57.1210μs 24.8667μs 40.2145 KOps/s 40.8659 KOps/s $\color{#d91a1a}-1.59\%$
test_step_mdp_speed[True-True-True-False-True] 52.4510μs 25.1448μs 39.7696 KOps/s 40.4715 KOps/s $\color{#d91a1a}-1.73\%$
test_step_mdp_speed[True-True-True-False-False] 46.1010μs 13.5718μs 73.6824 KOps/s 73.3074 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-True-False-True-True] 80.8020μs 46.4911μs 21.5095 KOps/s 21.3150 KOps/s $\color{#35bf28}+0.91\%$
test_step_mdp_speed[True-True-False-True-False] 63.5410μs 27.0362μs 36.9875 KOps/s 36.6295 KOps/s $\color{#35bf28}+0.98\%$
test_step_mdp_speed[True-True-False-False-True] 53.1410μs 27.1492μs 36.8335 KOps/s 36.5110 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[True-True-False-False-False] 48.0810μs 16.2639μs 61.4858 KOps/s 60.8334 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[True-False-True-True-True] 84.6110μs 49.9677μs 20.0129 KOps/s 20.1509 KOps/s $\color{#d91a1a}-0.68\%$
test_step_mdp_speed[True-False-True-True-False] 57.5110μs 29.9095μs 33.4342 KOps/s 33.0524 KOps/s $\color{#35bf28}+1.16\%$
test_step_mdp_speed[True-False-True-False-True] 58.8610μs 27.3706μs 36.5355 KOps/s 36.5809 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-False-True-False-False] 49.2810μs 16.1535μs 61.9061 KOps/s 61.9256 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[True-False-False-True-True] 81.2520μs 52.3347μs 19.1078 KOps/s 19.1482 KOps/s $\color{#d91a1a}-0.21\%$
test_step_mdp_speed[True-False-False-True-False] 65.1720μs 32.4669μs 30.8006 KOps/s 30.9152 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-False-False-False-True] 55.2210μs 29.5863μs 33.7994 KOps/s 33.4470 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-False-False-False-False] 44.9010μs 18.7424μs 53.3550 KOps/s 53.1412 KOps/s $\color{#35bf28}+0.40\%$
test_step_mdp_speed[False-True-True-True-True] 81.2720μs 49.0572μs 20.3844 KOps/s 20.2202 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[False-True-True-True-False] 60.0010μs 29.6961μs 33.6744 KOps/s 33.4762 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[False-True-True-False-True] 2.3893ms 31.5456μs 31.7001 KOps/s 31.6349 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[False-True-True-False-False] 48.3300μs 18.6811μs 53.5300 KOps/s 54.5083 KOps/s $\color{#d91a1a}-1.79\%$
test_step_mdp_speed[False-True-False-True-True] 81.0520μs 52.5982μs 19.0121 KOps/s 18.8814 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[False-True-False-True-False] 63.7510μs 32.6416μs 30.6357 KOps/s 30.1596 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[False-True-False-False-True] 66.7110μs 34.1165μs 29.3113 KOps/s 29.5008 KOps/s $\color{#d91a1a}-0.64\%$
test_step_mdp_speed[False-True-False-False-False] 49.7210μs 21.1906μs 47.1908 KOps/s 47.8490 KOps/s $\color{#d91a1a}-1.38\%$
test_step_mdp_speed[False-False-True-True-True] 88.4220μs 54.9471μs 18.1993 KOps/s 18.5070 KOps/s $\color{#d91a1a}-1.66\%$
test_step_mdp_speed[False-False-True-True-False] 84.7010μs 35.0695μs 28.5148 KOps/s 28.4415 KOps/s $\color{#35bf28}+0.26\%$
test_step_mdp_speed[False-False-True-False-True] 63.5910μs 33.1814μs 30.1373 KOps/s 29.5528 KOps/s $\color{#35bf28}+1.98\%$
test_step_mdp_speed[False-False-True-False-False] 56.2810μs 21.1135μs 47.3631 KOps/s 47.9190 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[False-False-False-True-True] 82.0520μs 55.9648μs 17.8684 KOps/s 17.8507 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[False-False-False-True-False] 70.7910μs 37.6521μs 26.5589 KOps/s 26.6685 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[False-False-False-False-True] 65.6810μs 35.4167μs 28.2353 KOps/s 28.2795 KOps/s $\color{#d91a1a}-0.16\%$
test_step_mdp_speed[False-False-False-False-False] 51.3310μs 22.9255μs 43.6196 KOps/s 43.8192 KOps/s $\color{#d91a1a}-0.46\%$
test_values[generalized_advantage_estimate-True-True] 10.6868ms 10.4478ms 95.7136 Ops/s 95.4198 Ops/s $\color{#35bf28}+0.31\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.3061ms 17.5343ms 57.0311 Ops/s 88.8184 Ops/s $\textbf{\color{#d91a1a}-35.79\%}$
test_values[td0_return_estimate-False-False] 0.2030ms 0.1209ms 8.2707 KOps/s 7.4650 KOps/s $\textbf{\color{#35bf28}+10.79\%}$
test_values[td1_return_estimate-False-False] 27.9313ms 27.4454ms 36.4360 Ops/s 37.1041 Ops/s $\color{#d91a1a}-1.80\%$
test_values[vec_td1_return_estimate-False-False] 19.7703ms 17.6748ms 56.5776 Ops/s 88.2102 Ops/s $\textbf{\color{#d91a1a}-35.86\%}$
test_values[td_lambda_return_estimate-True-False] 40.8586ms 40.1466ms 24.9087 Ops/s 25.3155 Ops/s $\color{#d91a1a}-1.61\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.5926ms 17.6319ms 56.7155 Ops/s 88.3215 Ops/s $\textbf{\color{#d91a1a}-35.79\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.6429ms 9.3414ms 107.0502 Ops/s 109.0642 Ops/s $\color{#d91a1a}-1.85\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7200ms 1.5090ms 662.6926 Ops/s 671.4270 Ops/s $\color{#d91a1a}-1.30\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7184ms 0.4057ms 2.4648 KOps/s 2.4791 KOps/s $\color{#d91a1a}-0.58\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.2977ms 33.7146ms 29.6607 Ops/s 40.6260 Ops/s $\textbf{\color{#d91a1a}-26.99\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.1257ms 1.7565ms 569.3169 Ops/s 574.7551 Ops/s $\color{#d91a1a}-0.95\%$
test_dqn_speed[False-None] 1.7395ms 1.3588ms 735.9225 Ops/s 738.2648 Ops/s $\color{#d91a1a}-0.32\%$
test_dqn_speed[False-backward] 1.9405ms 1.8558ms 538.8618 Ops/s 543.3892 Ops/s $\color{#d91a1a}-0.83\%$
test_dqn_speed[True-None] 0.7373ms 0.5290ms 1.8904 KOps/s 1.8224 KOps/s $\color{#35bf28}+3.73\%$
test_dqn_speed[True-backward] 1.0072ms 0.9534ms 1.0488 KOps/s 858.2918 Ops/s $\textbf{\color{#35bf28}+22.20\%}$
test_dqn_speed[reduce-overhead-None] 0.8946ms 0.5279ms 1.8943 KOps/s 1.7955 KOps/s $\textbf{\color{#35bf28}+5.50\%}$
test_dqn_speed[reduce-overhead-backward] 0.9933ms 0.9562ms 1.0459 KOps/s 1.0398 KOps/s $\color{#35bf28}+0.59\%$
test_ddpg_speed[False-None] 3.0395ms 2.7355ms 365.5624 Ops/s 361.9854 Ops/s $\color{#35bf28}+0.99\%$
test_ddpg_speed[False-backward] 4.2788ms 3.9002ms 256.3965 Ops/s 254.7970 Ops/s $\color{#35bf28}+0.63\%$
test_ddpg_speed[True-None] 1.7454ms 1.3961ms 716.3002 Ops/s 693.7104 Ops/s $\color{#35bf28}+3.26\%$
test_ddpg_speed[True-backward] 2.7249ms 2.3605ms 423.6469 Ops/s 403.4695 Ops/s $\textbf{\color{#35bf28}+5.00\%}$
test_ddpg_speed[reduce-overhead-None] 0.2332s 1.7042ms 586.7721 Ops/s 711.0032 Ops/s $\textbf{\color{#d91a1a}-17.47\%}$
test_ddpg_speed[reduce-overhead-backward] 2.3680ms 2.3248ms 430.1432 Ops/s 427.6945 Ops/s $\color{#35bf28}+0.57\%$
test_sac_speed[False-None] 7.8410ms 7.5073ms 133.2044 Ops/s 132.8071 Ops/s $\color{#35bf28}+0.30\%$
test_sac_speed[False-backward] 11.4043ms 10.6184ms 94.1765 Ops/s 62.5203 Ops/s $\textbf{\color{#35bf28}+50.63\%}$
test_sac_speed[True-None] 2.4434ms 2.1305ms 469.3778 Ops/s 461.1462 Ops/s $\color{#35bf28}+1.79\%$
test_sac_speed[True-backward] 4.1044ms 4.0050ms 249.6893 Ops/s 247.3983 Ops/s $\color{#35bf28}+0.93\%$
test_sac_speed[reduce-overhead-None] 2.4797ms 2.1366ms 468.0348 Ops/s 469.5932 Ops/s $\color{#d91a1a}-0.33\%$
test_sac_speed[reduce-overhead-backward] 4.1783ms 4.0117ms 249.2735 Ops/s 248.7836 Ops/s $\color{#35bf28}+0.20\%$
test_redq_speed[False-None] 11.5503ms 10.1435ms 98.5853 Ops/s 99.1487 Ops/s $\color{#d91a1a}-0.57\%$
test_redq_speed[False-backward] 21.7661ms 17.4508ms 57.3040 Ops/s 57.3424 Ops/s $\color{#d91a1a}-0.07\%$
test_redq_speed[True-None] 4.4900ms 4.2786ms 233.7199 Ops/s 226.5137 Ops/s $\color{#35bf28}+3.18\%$
test_redq_speed[True-backward] 10.0620ms 9.6726ms 103.3851 Ops/s 98.0504 Ops/s $\textbf{\color{#35bf28}+5.44\%}$
test_redq_speed[reduce-overhead-None] 5.4137ms 4.3635ms 229.1729 Ops/s 229.4995 Ops/s $\color{#d91a1a}-0.14\%$
test_redq_speed[reduce-overhead-backward] 10.1224ms 9.6596ms 103.5239 Ops/s 101.7661 Ops/s $\color{#35bf28}+1.73\%$
test_redq_deprec_speed[False-None] 10.8866ms 10.5014ms 95.2255 Ops/s 96.3880 Ops/s $\color{#d91a1a}-1.21\%$
test_redq_deprec_speed[False-backward] 15.5262ms 15.1321ms 66.0846 Ops/s 67.1328 Ops/s $\color{#d91a1a}-1.56\%$
test_redq_deprec_speed[True-None] 3.8601ms 3.5567ms 281.1593 Ops/s 276.6961 Ops/s $\color{#35bf28}+1.61\%$
test_redq_deprec_speed[True-backward] 7.4864ms 7.2846ms 137.2768 Ops/s 128.0330 Ops/s $\textbf{\color{#35bf28}+7.22\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.9574ms 3.5543ms 281.3496 Ops/s 280.9025 Ops/s $\color{#35bf28}+0.16\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.6653ms 7.3022ms 136.9448 Ops/s 137.7450 Ops/s $\color{#d91a1a}-0.58\%$
test_td3_speed[False-None] 7.5778ms 7.4883ms 133.5419 Ops/s 131.6178 Ops/s $\color{#35bf28}+1.46\%$
test_td3_speed[False-backward] 10.6884ms 10.2534ms 97.5288 Ops/s 96.5438 Ops/s $\color{#35bf28}+1.02\%$
test_td3_speed[True-None] 1.7753ms 1.7486ms 571.8993 Ops/s 561.0861 Ops/s $\color{#35bf28}+1.93\%$
test_td3_speed[True-backward] 3.6530ms 3.5589ms 280.9874 Ops/s 245.3296 Ops/s $\textbf{\color{#35bf28}+14.53\%}$
test_td3_speed[reduce-overhead-None] 1.8431ms 1.7711ms 564.6159 Ops/s 561.9824 Ops/s $\color{#35bf28}+0.47\%$
test_td3_speed[reduce-overhead-backward] 3.6000ms 3.5130ms 284.6559 Ops/s 283.5418 Ops/s $\color{#35bf28}+0.39\%$
test_cql_speed[False-None] 0.2825s 31.9277ms 31.3207 Ops/s 39.7213 Ops/s $\textbf{\color{#d91a1a}-21.15\%}$
test_cql_speed[False-backward] 37.6770ms 34.1453ms 29.2866 Ops/s 28.7225 Ops/s $\color{#35bf28}+1.96\%$
test_cql_speed[True-None] 12.2929ms 12.0401ms 83.0555 Ops/s 82.0621 Ops/s $\color{#35bf28}+1.21\%$
test_cql_speed[True-backward] 18.2906ms 17.9896ms 55.5876 Ops/s 55.5872 Ops/s $+0.00\%$
test_cql_speed[reduce-overhead-None] 12.4585ms 12.1420ms 82.3588 Ops/s 83.4085 Ops/s $\color{#d91a1a}-1.26\%$
test_cql_speed[reduce-overhead-backward] 18.4004ms 18.0378ms 55.4392 Ops/s 56.9552 Ops/s $\color{#d91a1a}-2.66\%$
test_a2c_speed[False-None] 5.5482ms 5.2585ms 190.1678 Ops/s 189.2264 Ops/s $\color{#35bf28}+0.50\%$
test_a2c_speed[False-backward] 11.8004ms 11.4533ms 87.3112 Ops/s 86.3626 Ops/s $\color{#35bf28}+1.10\%$
test_a2c_speed[True-None] 4.0031ms 3.6595ms 273.2643 Ops/s 267.3450 Ops/s $\color{#35bf28}+2.21\%$
test_a2c_speed[True-backward] 11.5595ms 9.4292ms 106.0530 Ops/s 116.3655 Ops/s $\textbf{\color{#d91a1a}-8.86\%}$
test_a2c_speed[reduce-overhead-None] 3.8542ms 3.6832ms 271.5009 Ops/s 271.0683 Ops/s $\color{#35bf28}+0.16\%$
test_a2c_speed[reduce-overhead-backward] 8.6841ms 8.4213ms 118.7467 Ops/s 118.4282 Ops/s $\color{#35bf28}+0.27\%$
test_ppo_speed[False-None] 5.9161ms 5.6840ms 175.9331 Ops/s 172.6842 Ops/s $\color{#35bf28}+1.88\%$
test_ppo_speed[False-backward] 12.2896ms 11.9838ms 83.4456 Ops/s 82.4739 Ops/s $\color{#35bf28}+1.18\%$
test_ppo_speed[True-None] 4.0358ms 3.6524ms 273.7927 Ops/s 266.2545 Ops/s $\color{#35bf28}+2.83\%$
test_ppo_speed[True-backward] 8.5912ms 8.3766ms 119.3806 Ops/s 115.8088 Ops/s $\color{#35bf28}+3.08\%$
test_ppo_speed[reduce-overhead-None] 3.7687ms 3.6188ms 276.3334 Ops/s 275.7246 Ops/s $\color{#35bf28}+0.22\%$
test_ppo_speed[reduce-overhead-backward] 8.9906ms 8.3608ms 119.6057 Ops/s 118.7576 Ops/s $\color{#35bf28}+0.71\%$
test_reinforce_speed[False-None] 4.8102ms 4.3938ms 227.5942 Ops/s 227.6917 Ops/s $\color{#d91a1a}-0.04\%$
test_reinforce_speed[False-backward] 7.3074ms 7.0513ms 141.8185 Ops/s 138.7761 Ops/s $\color{#35bf28}+2.19\%$
test_reinforce_speed[True-None] 3.0115ms 2.8388ms 352.2586 Ops/s 341.6756 Ops/s $\color{#35bf28}+3.10\%$
test_reinforce_speed[True-backward] 7.8322ms 7.5900ms 131.7522 Ops/s 131.6245 Ops/s $\color{#35bf28}+0.10\%$
test_reinforce_speed[reduce-overhead-None] 3.5056ms 2.8564ms 350.0912 Ops/s 352.2927 Ops/s $\color{#d91a1a}-0.62\%$
test_reinforce_speed[reduce-overhead-backward] 7.8541ms 7.5830ms 131.8734 Ops/s 130.7672 Ops/s $\color{#35bf28}+0.85\%$
test_iql_speed[False-None] 24.3079ms 19.1905ms 52.1090 Ops/s 52.6967 Ops/s $\color{#d91a1a}-1.12\%$
test_iql_speed[False-backward] 35.4816ms 29.7648ms 33.5967 Ops/s 34.4474 Ops/s $\color{#d91a1a}-2.47\%$
test_iql_speed[True-None] 8.7911ms 8.4403ms 118.4793 Ops/s 115.7606 Ops/s $\color{#35bf28}+2.35\%$
test_iql_speed[True-backward] 16.8795ms 16.5932ms 60.2655 Ops/s 59.3805 Ops/s $\color{#35bf28}+1.49\%$
test_iql_speed[reduce-overhead-None] 8.8042ms 8.4991ms 117.6596 Ops/s 113.6488 Ops/s $\color{#35bf28}+3.53\%$
test_iql_speed[reduce-overhead-backward] 16.8061ms 16.4969ms 60.6176 Ops/s 60.7715 Ops/s $\color{#d91a1a}-0.25\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4680ms 6.0982ms 163.9818 Ops/s 162.8213 Ops/s $\color{#35bf28}+0.71\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6441ms 0.3396ms 2.9450 KOps/s 3.4537 KOps/s $\textbf{\color{#d91a1a}-14.73\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5538ms 0.3229ms 3.0965 KOps/s 3.3117 KOps/s $\textbf{\color{#d91a1a}-6.50\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3644ms 5.8621ms 170.5866 Ops/s 170.8459 Ops/s $\color{#d91a1a}-0.15\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9212ms 0.3327ms 3.0058 KOps/s 3.2693 KOps/s $\textbf{\color{#d91a1a}-8.06\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5260ms 0.3165ms 3.1597 KOps/s 3.2980 KOps/s $\color{#d91a1a}-4.19\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5809ms 1.3469ms 742.4517 Ops/s 802.2913 Ops/s $\textbf{\color{#d91a1a}-7.46\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5425ms 1.2628ms 791.8934 Ops/s 804.8838 Ops/s $\color{#d91a1a}-1.61\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1936ms 6.0173ms 166.1877 Ops/s 168.0836 Ops/s $\color{#d91a1a}-1.13\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9810ms 0.4487ms 2.2289 KOps/s 1.0901 KOps/s $\textbf{\color{#35bf28}+104.46\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6911ms 0.4163ms 2.4020 KOps/s 2.1552 KOps/s $\textbf{\color{#35bf28}+11.45\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 10.0129ms 6.0821ms 164.4174 Ops/s 167.4699 Ops/s $\color{#d91a1a}-1.82\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.6046ms 0.3060ms 3.2678 KOps/s 3.8887 KOps/s $\textbf{\color{#d91a1a}-15.97\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6453ms 0.2857ms 3.4997 KOps/s 4.1982 KOps/s $\textbf{\color{#d91a1a}-16.64\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0354ms 5.8241ms 171.7016 Ops/s 168.9975 Ops/s $\color{#35bf28}+1.60\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5808s 0.9703ms 1.0306 KOps/s 2.8879 KOps/s $\textbf{\color{#d91a1a}-64.31\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6232ms 0.3180ms 3.1446 KOps/s 3.0089 KOps/s $\color{#35bf28}+4.51\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1606ms 6.0717ms 164.6991 Ops/s 164.8955 Ops/s $\color{#d91a1a}-0.12\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.3388ms 0.4799ms 2.0838 KOps/s 2.3071 KOps/s $\textbf{\color{#d91a1a}-9.68\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7438ms 0.4618ms 2.1656 KOps/s 2.1702 KOps/s $\color{#d91a1a}-0.21\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0399ms 5.4419ms 183.7582 Ops/s 178.7364 Ops/s $\color{#35bf28}+2.81\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.3113ms 2.1599ms 462.9766 Ops/s 413.7985 Ops/s $\textbf{\color{#35bf28}+11.88\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.2741ms 1.0799ms 926.0108 Ops/s 882.0573 Ops/s $\color{#35bf28}+4.98\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4367s 14.0912ms 70.9664 Ops/s 60.7848 Ops/s $\textbf{\color{#35bf28}+16.75\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.7453ms 1.7791ms 562.0856 Ops/s 517.0500 Ops/s $\textbf{\color{#35bf28}+8.71\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.1645ms 1.2265ms 815.3304 Ops/s 777.1859 Ops/s $\color{#35bf28}+4.91\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.6404ms 5.6572ms 176.7658 Ops/s 172.4958 Ops/s $\color{#35bf28}+2.48\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.1541ms 2.1851ms 457.6482 Ops/s 455.5783 Ops/s $\color{#35bf28}+0.45\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.1986ms 1.3880ms 720.4699 Ops/s 713.5175 Ops/s $\color{#35bf28}+0.97\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 58.4822ms 57.2294ms 17.4735 Ops/s 16.0495 Ops/s $\textbf{\color{#35bf28}+8.87\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 17.9482ms 16.4854ms 60.6599 Ops/s 60.7904 Ops/s $\color{#d91a1a}-0.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 58.4271ms 57.4206ms 17.4154 Ops/s 17.2439 Ops/s $\color{#35bf28}+0.99\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.6040ms 16.7554ms 59.6822 Ops/s 60.2282 Ops/s $\color{#d91a1a}-0.91\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 58.3035ms 57.0946ms 17.5148 Ops/s 16.9787 Ops/s $\color{#35bf28}+3.16\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.2854ms 18.0606ms 55.3690 Ops/s 54.6938 Ops/s $\color{#35bf28}+1.23\%$

@vmoens vmoens merged commit 8ac73aa into main Jul 24, 2025
58 of 71 checks passed
@vmoens vmoens deleted the fix-benches branch July 24, 2025 16:13
@github-actions
Copy link

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7626s 0.7619s 1.3125 Ops/s 1.2778 Ops/s $\color{#35bf28}+2.72\%$
test_transformed 1.4577s 1.3802s 0.7245 Ops/s 0.7200 Ops/s $\color{#35bf28}+0.63\%$
test_serial 2.2411s 2.2367s 0.4471 Ops/s 0.4453 Ops/s $\color{#35bf28}+0.41\%$
test_parallel 1.9600s 1.8722s 0.5341 Ops/s 0.5354 Ops/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[True-True-True-True-True] 0.1966ms 42.8011μs 23.3639 KOps/s 23.2868 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[True-True-True-True-False] 0.1064ms 24.4041μs 40.9767 KOps/s 41.2740 KOps/s $\color{#d91a1a}-0.72\%$
test_step_mdp_speed[True-True-True-False-True] 0.1074ms 23.8442μs 41.9389 KOps/s 41.5466 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[True-True-True-False-False] 45.3910μs 13.3998μs 74.6280 KOps/s 74.8986 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[True-True-False-True-True] 74.1620μs 44.6150μs 22.4140 KOps/s 22.3406 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[True-True-False-True-False] 90.9720μs 26.7480μs 37.3860 KOps/s 38.3470 KOps/s $\color{#d91a1a}-2.51\%$
test_step_mdp_speed[True-True-False-False-True] 0.1930ms 26.3475μs 37.9542 KOps/s 38.1321 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-True-False-False-False] 43.6710μs 15.6991μs 63.6978 KOps/s 64.4296 KOps/s $\color{#d91a1a}-1.14\%$
test_step_mdp_speed[True-False-True-True-True] 82.2020μs 47.9283μs 20.8645 KOps/s 20.9761 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[True-False-True-True-False] 0.1217ms 29.2596μs 34.1768 KOps/s 34.4046 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[True-False-True-False-True] 0.2040ms 26.5832μs 37.6178 KOps/s 37.5891 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[True-False-True-False-False] 48.8610μs 15.9974μs 62.5100 KOps/s 63.0224 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[True-False-False-True-True] 80.7020μs 50.5262μs 19.7917 KOps/s 20.0272 KOps/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-False-False-True-False] 62.8210μs 31.6168μs 31.6288 KOps/s 32.1776 KOps/s $\color{#d91a1a}-1.71\%$
test_step_mdp_speed[True-False-False-False-True] 76.8320μs 29.1316μs 34.3270 KOps/s 34.5416 KOps/s $\color{#d91a1a}-0.62\%$
test_step_mdp_speed[True-False-False-False-False] 41.9310μs 18.6640μs 53.5790 KOps/s 54.4802 KOps/s $\color{#d91a1a}-1.65\%$
test_step_mdp_speed[False-True-True-True-True] 91.8810μs 47.7473μs 20.9436 KOps/s 21.2352 KOps/s $\color{#d91a1a}-1.37\%$
test_step_mdp_speed[False-True-True-True-False] 0.1482ms 29.0263μs 34.4515 KOps/s 35.0057 KOps/s $\color{#d91a1a}-1.58\%$
test_step_mdp_speed[False-True-True-False-True] 2.5373ms 31.2256μs 32.0250 KOps/s 33.2458 KOps/s $\color{#d91a1a}-3.67\%$
test_step_mdp_speed[False-True-True-False-False] 55.7010μs 18.1698μs 55.0363 KOps/s 56.0943 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[False-True-False-True-True] 83.5420μs 50.9230μs 19.6375 KOps/s 19.7532 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[False-True-False-True-False] 62.0510μs 31.7737μs 31.4725 KOps/s 31.2940 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[False-True-False-False-True] 63.5610μs 32.2458μs 31.0117 KOps/s 30.3659 KOps/s $\color{#35bf28}+2.13\%$
test_step_mdp_speed[False-True-False-False-False] 61.2520μs 20.1268μs 49.6851 KOps/s 49.9615 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-False-True-True-True] 0.1780ms 52.8632μs 18.9167 KOps/s 18.6193 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[False-False-True-True-False] 59.6310μs 33.9832μs 29.4263 KOps/s 29.4529 KOps/s $\color{#d91a1a}-0.09\%$
test_step_mdp_speed[False-False-True-False-True] 76.8510μs 32.4634μs 30.8040 KOps/s 30.4986 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[False-False-True-False-False] 46.6710μs 19.8678μs 50.3327 KOps/s 49.9845 KOps/s $\color{#35bf28}+0.70\%$
test_step_mdp_speed[False-False-False-True-True] 88.2020μs 54.6401μs 18.3016 KOps/s 18.5163 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[False-False-False-True-False] 73.6110μs 36.1528μs 27.6604 KOps/s 27.3961 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[False-False-False-False-True] 80.6120μs 34.3515μs 29.1108 KOps/s 29.0548 KOps/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[False-False-False-False-False] 90.5410μs 22.1047μs 45.2393 KOps/s 44.6361 KOps/s $\color{#35bf28}+1.35\%$
test_values[generalized_advantage_estimate-True-True] 21.3690ms 20.8634ms 47.9308 Ops/s 47.8098 Ops/s $\color{#35bf28}+0.25\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1365s 3.6372ms 274.9405 Ops/s 276.7853 Ops/s $\color{#d91a1a}-0.67\%$
test_values[td0_return_estimate-False-False] 0.1060ms 79.5247μs 12.5747 KOps/s 12.9129 KOps/s $\color{#d91a1a}-2.62\%$
test_values[td1_return_estimate-False-False] 49.9670ms 49.1655ms 20.3395 Ops/s 20.3890 Ops/s $\color{#d91a1a}-0.24\%$
test_values[vec_td1_return_estimate-False-False] 1.3081ms 1.0972ms 911.3859 Ops/s 914.1158 Ops/s $\color{#d91a1a}-0.30\%$
test_values[td_lambda_return_estimate-True-False] 81.9671ms 80.0300ms 12.4953 Ops/s 12.4598 Ops/s $\color{#35bf28}+0.29\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2570ms 1.0919ms 915.8456 Ops/s 913.3731 Ops/s $\color{#35bf28}+0.27\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 21.2490ms 21.0015ms 47.6156 Ops/s 47.7638 Ops/s $\color{#d91a1a}-0.31\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0507ms 0.7420ms 1.3477 KOps/s 1.3415 KOps/s $\color{#35bf28}+0.46\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8360ms 0.6653ms 1.5030 KOps/s 1.4961 KOps/s $\color{#35bf28}+0.46\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6244ms 1.4787ms 676.2762 Ops/s 676.5796 Ops/s $\color{#d91a1a}-0.04\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8695ms 0.6834ms 1.4633 KOps/s 1.4617 KOps/s $\color{#35bf28}+0.11\%$
test_dqn_speed[False-None] 1.6923ms 1.4958ms 668.5575 Ops/s 661.6773 Ops/s $\color{#35bf28}+1.04\%$
test_dqn_speed[False-backward] 2.2040ms 2.1608ms 462.7846 Ops/s 463.7272 Ops/s $\color{#d91a1a}-0.20\%$
test_dqn_speed[True-None] 1.1153ms 0.5789ms 1.7275 KOps/s 1.7339 KOps/s $\color{#d91a1a}-0.37\%$
test_dqn_speed[True-backward] 1.2402ms 1.1759ms 850.4408 Ops/s 842.7513 Ops/s $\color{#35bf28}+0.91\%$
test_dqn_speed[reduce-overhead-None] 1.0072ms 0.6007ms 1.6647 KOps/s 1.6126 KOps/s $\color{#35bf28}+3.23\%$
test_dqn_speed[reduce-overhead-backward] 1.1277ms 1.0061ms 993.9678 Ops/s 924.0835 Ops/s $\textbf{\color{#35bf28}+7.56\%}$
test_ddpg_speed[False-None] 3.2602ms 2.8700ms 348.4372 Ops/s 352.2776 Ops/s $\color{#d91a1a}-1.09\%$
test_ddpg_speed[False-backward] 4.3138ms 4.2078ms 237.6553 Ops/s 230.6908 Ops/s $\color{#35bf28}+3.02\%$
test_ddpg_speed[True-None] 1.7817ms 1.3840ms 722.5193 Ops/s 722.1215 Ops/s $\color{#35bf28}+0.06\%$
test_ddpg_speed[True-backward] 2.7920ms 2.6091ms 383.2808 Ops/s 385.4013 Ops/s $\color{#d91a1a}-0.55\%$
test_ddpg_speed[reduce-overhead-None] 1.6093ms 1.4000ms 714.2672 Ops/s 703.1165 Ops/s $\color{#35bf28}+1.59\%$
test_ddpg_speed[reduce-overhead-backward] 0.1953s 0.1888s 5.2968 Ops/s 4.2368 Ops/s $\textbf{\color{#35bf28}+25.02\%}$
test_sac_speed[False-None] 8.4569ms 8.0530ms 124.1773 Ops/s 124.0079 Ops/s $\color{#35bf28}+0.14\%$
test_sac_speed[False-backward] 11.9254ms 11.2072ms 89.2282 Ops/s 89.5558 Ops/s $\color{#d91a1a}-0.37\%$
test_sac_speed[True-None] 2.5692ms 1.9357ms 516.6026 Ops/s 520.8488 Ops/s $\color{#d91a1a}-0.82\%$
test_sac_speed[True-backward] 3.9330ms 3.7717ms 265.1290 Ops/s 262.8297 Ops/s $\color{#35bf28}+0.87\%$
test_sac_speed[reduce-overhead-None] 19.7809ms 11.3725ms 87.9311 Ops/s 87.5759 Ops/s $\color{#35bf28}+0.41\%$
test_sac_speed[reduce-overhead-backward] 1.9272ms 1.8022ms 554.8663 Ops/s 602.3754 Ops/s $\textbf{\color{#d91a1a}-7.89\%}$
test_redq_deprec_speed[False-None] 9.3492ms 8.9942ms 111.1832 Ops/s 110.3926 Ops/s $\color{#35bf28}+0.72\%$
test_redq_deprec_speed[False-backward] 12.9923ms 12.5011ms 79.9930 Ops/s 80.4516 Ops/s $\color{#d91a1a}-0.57\%$
test_redq_deprec_speed[True-None] 2.7415ms 2.5742ms 388.4708 Ops/s 387.6351 Ops/s $\color{#35bf28}+0.22\%$
test_redq_deprec_speed[True-backward] 4.9342ms 4.5866ms 218.0271 Ops/s 225.7458 Ops/s $\color{#d91a1a}-3.42\%$
test_redq_deprec_speed[reduce-overhead-None] 2.7375ms 2.5600ms 390.6304 Ops/s 384.1544 Ops/s $\color{#35bf28}+1.69\%$
test_redq_deprec_speed[reduce-overhead-backward] 5.0009ms 4.6139ms 216.7341 Ops/s 225.2878 Ops/s $\color{#d91a1a}-3.80\%$
test_td3_speed[False-None] 7.9840ms 7.9276ms 126.1412 Ops/s 118.7885 Ops/s $\textbf{\color{#35bf28}+6.19\%}$
test_td3_speed[False-backward] 11.3245ms 10.6694ms 93.7259 Ops/s 94.8448 Ops/s $\color{#d91a1a}-1.18\%$
test_td3_speed[True-None] 1.7609ms 1.7329ms 577.0558 Ops/s 573.7423 Ops/s $\color{#35bf28}+0.58\%$
test_td3_speed[True-backward] 3.7454ms 3.5630ms 280.6632 Ops/s 288.2435 Ops/s $\color{#d91a1a}-2.63\%$
test_td3_speed[reduce-overhead-None] 49.1975ms 25.2970ms 39.5304 Ops/s 39.2017 Ops/s $\color{#35bf28}+0.84\%$
test_td3_speed[reduce-overhead-backward] 1.6560ms 1.4845ms 673.6500 Ops/s 743.5349 Ops/s $\textbf{\color{#d91a1a}-9.40\%}$
test_cql_speed[False-None] 17.0799ms 16.7372ms 59.7472 Ops/s 59.7157 Ops/s $\color{#35bf28}+0.05\%$
test_cql_speed[False-backward] 23.1053ms 22.5578ms 44.3305 Ops/s 45.0621 Ops/s $\color{#d91a1a}-1.62\%$
test_cql_speed[True-None] 3.6014ms 3.4311ms 291.4554 Ops/s 288.4590 Ops/s $\color{#35bf28}+1.04\%$
test_cql_speed[True-backward] 6.3675ms 5.9596ms 167.7960 Ops/s 165.9439 Ops/s $\color{#35bf28}+1.12\%$
test_cql_speed[reduce-overhead-None] 19.3747ms 12.2936ms 81.3434 Ops/s 79.4833 Ops/s $\color{#35bf28}+2.34\%$
test_cql_speed[reduce-overhead-backward] 2.0873ms 1.9610ms 509.9471 Ops/s 501.2784 Ops/s $\color{#35bf28}+1.73\%$
test_a2c_speed[False-None] 3.3049ms 3.1535ms 317.1048 Ops/s 315.7353 Ops/s $\color{#35bf28}+0.43\%$
test_a2c_speed[False-backward] 7.4332ms 6.4933ms 154.0047 Ops/s 153.9630 Ops/s $\color{#35bf28}+0.03\%$
test_a2c_speed[True-None] 1.5504ms 1.3631ms 733.6460 Ops/s 747.1627 Ops/s $\color{#d91a1a}-1.81\%$
test_a2c_speed[True-backward] 3.2788ms 3.2217ms 310.3935 Ops/s 305.4410 Ops/s $\color{#35bf28}+1.62\%$
test_a2c_speed[reduce-overhead-None] 14.5798ms 8.1633ms 122.4991 Ops/s 116.2440 Ops/s $\textbf{\color{#35bf28}+5.38\%}$
test_a2c_speed[reduce-overhead-backward] 1.7308ms 1.6170ms 618.4368 Ops/s 614.3896 Ops/s $\color{#35bf28}+0.66\%$
test_ppo_speed[False-None] 4.0251ms 3.7557ms 266.2593 Ops/s 265.8258 Ops/s $\color{#35bf28}+0.16\%$
test_ppo_speed[False-backward] 7.6680ms 7.2272ms 138.3654 Ops/s 137.2651 Ops/s $\color{#35bf28}+0.80\%$
test_ppo_speed[True-None] 1.5999ms 1.4417ms 693.6050 Ops/s 678.8736 Ops/s $\color{#35bf28}+2.17\%$
test_ppo_speed[True-backward] 3.5542ms 3.4162ms 292.7253 Ops/s 292.1083 Ops/s $\color{#35bf28}+0.21\%$
test_ppo_speed[reduce-overhead-None] 1.5738ms 1.4305ms 699.0744 Ops/s 687.8699 Ops/s $\color{#35bf28}+1.63\%$
test_ppo_speed[reduce-overhead-backward] 3.5469ms 3.4021ms 293.9381 Ops/s 286.3186 Ops/s $\color{#35bf28}+2.66\%$
test_reinforce_speed[False-None] 2.4258ms 2.2548ms 443.4984 Ops/s 439.4975 Ops/s $\color{#35bf28}+0.91\%$
test_reinforce_speed[False-backward] 3.4762ms 3.4243ms 292.0267 Ops/s 291.1047 Ops/s $\color{#35bf28}+0.32\%$
test_reinforce_speed[True-None] 1.4808ms 1.3094ms 763.6906 Ops/s 764.7599 Ops/s $\color{#d91a1a}-0.14\%$
test_reinforce_speed[True-backward] 3.3668ms 3.2006ms 312.4447 Ops/s 307.7258 Ops/s $\color{#35bf28}+1.53\%$
test_reinforce_speed[reduce-overhead-None] 19.3033ms 10.2889ms 97.1925 Ops/s 98.0996 Ops/s $\color{#d91a1a}-0.92\%$
test_reinforce_speed[reduce-overhead-backward] 1.7059ms 1.6470ms 607.1607 Ops/s 605.5188 Ops/s $\color{#35bf28}+0.27\%$
test_iql_speed[False-None] 9.9406ms 9.2648ms 107.9353 Ops/s 108.4030 Ops/s $\color{#d91a1a}-0.43\%$
test_iql_speed[False-backward] 13.8526ms 13.4977ms 74.0869 Ops/s 74.1890 Ops/s $\color{#d91a1a}-0.14\%$
test_iql_speed[True-None] 2.5115ms 2.3114ms 432.6355 Ops/s 433.0994 Ops/s $\color{#d91a1a}-0.11\%$
test_iql_speed[True-backward] 5.8209ms 5.2020ms 192.2344 Ops/s 189.8274 Ops/s $\color{#35bf28}+1.27\%$
test_iql_speed[reduce-overhead-None] 17.7297ms 10.5882ms 94.4450 Ops/s 94.2548 Ops/s $\color{#35bf28}+0.20\%$
test_iql_speed[reduce-overhead-backward] 2.2224ms 2.0801ms 480.7498 Ops/s 497.3738 Ops/s $\color{#d91a1a}-3.34\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4352ms 6.0794ms 164.4905 Ops/s 163.6227 Ops/s $\color{#35bf28}+0.53\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5328ms 0.2639ms 3.7893 KOps/s 2.7679 KOps/s $\textbf{\color{#35bf28}+36.90\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5886ms 0.3283ms 3.0463 KOps/s 2.9040 KOps/s $\color{#35bf28}+4.90\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3687ms 5.8363ms 171.3407 Ops/s 169.9207 Ops/s $\color{#35bf28}+0.84\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7780ms 0.3042ms 3.2870 KOps/s 3.3335 KOps/s $\color{#d91a1a}-1.39\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7533ms 0.3319ms 3.0128 KOps/s 3.5675 KOps/s $\textbf{\color{#d91a1a}-15.55\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6384ms 1.3926ms 718.1052 Ops/s 789.8272 Ops/s $\textbf{\color{#d91a1a}-9.08\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5836ms 1.3302ms 751.7861 Ops/s 861.3300 Ops/s $\textbf{\color{#d91a1a}-12.72\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2594ms 6.0260ms 165.9463 Ops/s 165.0527 Ops/s $\color{#35bf28}+0.54\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9926ms 0.4433ms 2.2557 KOps/s 2.2180 KOps/s $\color{#35bf28}+1.70\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6630ms 0.4102ms 2.4376 KOps/s 2.3915 KOps/s $\color{#35bf28}+1.93\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9484ms 5.7357ms 174.3456 Ops/s 170.4000 Ops/s $\color{#35bf28}+2.32\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.7814ms 0.2660ms 3.7594 KOps/s 2.8423 KOps/s $\textbf{\color{#35bf28}+32.27\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4533ms 0.2452ms 4.0786 KOps/s 3.0001 KOps/s $\textbf{\color{#35bf28}+35.95\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0560ms 5.7099ms 175.1335 Ops/s 173.0162 Ops/s $\color{#35bf28}+1.22\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8676ms 0.3500ms 2.8574 KOps/s 3.0701 KOps/s $\textbf{\color{#d91a1a}-6.93\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5317ms 0.3040ms 3.2895 KOps/s 3.9021 KOps/s $\textbf{\color{#d91a1a}-15.70\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0997ms 5.9326ms 168.5612 Ops/s 168.2745 Ops/s $\color{#35bf28}+0.17\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9042ms 0.4882ms 2.0482 KOps/s 2.1263 KOps/s $\color{#d91a1a}-3.68\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7409ms 0.4662ms 2.1449 KOps/s 2.3432 KOps/s $\textbf{\color{#d91a1a}-8.46\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0056ms 5.4673ms 182.9063 Ops/s 49.9534 Ops/s $\textbf{\color{#35bf28}+266.15\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.2452ms 2.1007ms 476.0358 Ops/s 470.7256 Ops/s $\color{#35bf28}+1.13\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.9369ms 1.2838ms 778.9383 Ops/s 802.8780 Ops/s $\color{#d91a1a}-2.98\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.4926ms 5.6127ms 178.1658 Ops/s 179.5251 Ops/s $\color{#d91a1a}-0.76\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.7002ms 2.0711ms 482.8336 Ops/s 473.4055 Ops/s $\color{#35bf28}+1.99\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1598ms 1.2985ms 770.1098 Ops/s 820.9697 Ops/s $\textbf{\color{#d91a1a}-6.20\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5306s 16.2741ms 61.4474 Ops/s 171.7604 Ops/s $\textbf{\color{#d91a1a}-64.22\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.8172ms 2.2874ms 437.1869 Ops/s 435.0757 Ops/s $\color{#35bf28}+0.49\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.1263ms 1.3583ms 736.1971 Ops/s 752.9891 Ops/s $\color{#d91a1a}-2.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 62.7299ms 57.2908ms 17.4548 Ops/s 17.3129 Ops/s $\color{#35bf28}+0.82\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.0238ms 16.6435ms 60.0837 Ops/s 60.4723 Ops/s $\color{#d91a1a}-0.64\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 62.2849ms 59.1663ms 16.9015 Ops/s 17.2912 Ops/s $\color{#d91a1a}-2.25\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.1096ms 16.7652ms 59.6475 Ops/s 59.5698 Ops/s $\color{#35bf28}+0.13\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 61.3477ms 58.2782ms 17.1591 Ops/s 16.7886 Ops/s $\color{#35bf28}+2.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.3983ms 18.0693ms 55.3424 Ops/s 55.1895 Ops/s $\color{#35bf28}+0.28\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Benchmarks rl/benchmark changes 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.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants