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

[CI] Fix nightly build #2666

Merged
merged 1 commit into from
Dec 19, 2024
Merged

[CI] Fix nightly build #2666

merged 1 commit into from
Dec 19, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Dec 19, 2024

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Dec 19, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2666

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

✅ You can merge normally! (8 Unrelated Failures)

As of commit 917efc8 with merge base 1ce25f1 (image):

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

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

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

vmoens added a commit that referenced this pull request Dec 19, 2024
ghstack-source-id: 5502fa94b6abcc154e020dcb165093fdc30ca025
Pull Request resolved: #2666
@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 Dec 19, 2024
@vmoens vmoens merged commit 917efc8 into gh/vmoens/59/base Dec 19, 2024
88 of 92 checks passed
vmoens added a commit that referenced this pull request Dec 19, 2024
ghstack-source-id: 5502fa94b6abcc154e020dcb165093fdc30ca025
Pull Request resolved: #2666
@vmoens vmoens deleted the gh/vmoens/59/head branch December 19, 2024 10:39
Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}26$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4363s 0.4345s 2.3014 Ops/s 2.2412 Ops/s $\color{#35bf28}+2.68\%$
test_transformed 0.6103s 0.6090s 1.6420 Ops/s 1.6569 Ops/s $\color{#d91a1a}-0.90\%$
test_serial 1.3680s 1.3575s 0.7366 Ops/s 0.7292 Ops/s $\color{#35bf28}+1.02\%$
test_parallel 1.4267s 1.3342s 0.7495 Ops/s 0.7421 Ops/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[True-True-True-True-True] 0.1864ms 30.9442μs 32.3162 KOps/s 32.7891 KOps/s $\color{#d91a1a}-1.44\%$
test_step_mdp_speed[True-True-True-True-False] 49.1020μs 18.1513μs 55.0925 KOps/s 55.8368 KOps/s $\color{#d91a1a}-1.33\%$
test_step_mdp_speed[True-True-True-False-True] 60.3030μs 17.9142μs 55.8218 KOps/s 57.6780 KOps/s $\color{#d91a1a}-3.22\%$
test_step_mdp_speed[True-True-True-False-False] 57.6960μs 10.2370μs 97.6853 KOps/s 99.6311 KOps/s $\color{#d91a1a}-1.95\%$
test_step_mdp_speed[True-True-False-True-True] 67.8370μs 33.6284μs 29.7368 KOps/s 30.2979 KOps/s $\color{#d91a1a}-1.85\%$
test_step_mdp_speed[True-True-False-True-False] 52.0670μs 20.3643μs 49.1056 KOps/s 49.3011 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[True-True-False-False-True] 61.2350μs 19.7158μs 50.7206 KOps/s 50.7389 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[True-True-False-False-False] 53.7810μs 12.2208μs 81.8278 KOps/s 82.2420 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-False-True-True-True] 88.7160μs 35.7785μs 27.9497 KOps/s 28.7624 KOps/s $\color{#d91a1a}-2.83\%$
test_step_mdp_speed[True-False-True-True-False] 58.6600μs 22.0884μs 45.2727 KOps/s 45.0190 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-False-True-False-True] 0.1033ms 19.4265μs 51.4760 KOps/s 51.7930 KOps/s $\color{#d91a1a}-0.61\%$
test_step_mdp_speed[True-False-True-False-False] 38.7320μs 12.1827μs 82.0834 KOps/s 83.4121 KOps/s $\color{#d91a1a}-1.59\%$
test_step_mdp_speed[True-False-False-True-True] 75.2310μs 37.6477μs 26.5621 KOps/s 27.1251 KOps/s $\color{#d91a1a}-2.08\%$
test_step_mdp_speed[True-False-False-True-False] 52.0680μs 24.0737μs 41.5391 KOps/s 42.0170 KOps/s $\color{#d91a1a}-1.14\%$
test_step_mdp_speed[True-False-False-False-True] 49.0520μs 21.5055μs 46.4998 KOps/s 47.1559 KOps/s $\color{#d91a1a}-1.39\%$
test_step_mdp_speed[True-False-False-False-False] 36.8590μs 14.1097μs 70.8734 KOps/s 71.9291 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-True-True-True-True] 70.6020μs 36.0238μs 27.7595 KOps/s 28.6529 KOps/s $\color{#d91a1a}-3.12\%$
test_step_mdp_speed[False-True-True-True-False] 53.6300μs 22.4480μs 44.5474 KOps/s 45.3049 KOps/s $\color{#d91a1a}-1.67\%$
test_step_mdp_speed[False-True-True-False-True] 69.4100μs 22.7119μs 44.0298 KOps/s 44.7907 KOps/s $\color{#d91a1a}-1.70\%$
test_step_mdp_speed[False-True-True-False-False] 54.7030μs 13.6825μs 73.0860 KOps/s 73.3204 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[False-True-False-True-True] 0.1001ms 37.4239μs 26.7209 KOps/s 27.1586 KOps/s $\color{#d91a1a}-1.61\%$
test_step_mdp_speed[False-True-False-True-False] 47.9900μs 23.6539μs 42.2763 KOps/s 41.6820 KOps/s $\color{#35bf28}+1.43\%$
test_step_mdp_speed[False-True-False-False-True] 2.6274ms 24.3951μs 40.9919 KOps/s 40.7819 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[False-True-False-False-False] 50.5740μs 15.3042μs 65.3414 KOps/s 64.3856 KOps/s $\color{#35bf28}+1.48\%$
test_step_mdp_speed[False-False-True-True-True] 87.6450μs 39.0359μs 25.6174 KOps/s 25.9303 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[False-False-True-True-False] 51.5070μs 25.9356μs 38.5571 KOps/s 38.3457 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-False-True-False-True] 67.5160μs 24.5664μs 40.7060 KOps/s 41.7474 KOps/s $\color{#d91a1a}-2.49\%$
test_step_mdp_speed[False-False-True-False-False] 39.2540μs 15.6055μs 64.0802 KOps/s 65.5486 KOps/s $\color{#d91a1a}-2.24\%$
test_step_mdp_speed[False-False-False-True-True] 97.2220μs 41.0159μs 24.3808 KOps/s 24.9517 KOps/s $\color{#d91a1a}-2.29\%$
test_step_mdp_speed[False-False-False-True-False] 57.5770μs 27.9331μs 35.7998 KOps/s 36.2371 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[False-False-False-False-True] 53.4800μs 25.7504μs 38.8343 KOps/s 38.8380 KOps/s $-0.01\%$
test_step_mdp_speed[False-False-False-False-False] 52.6990μs 17.1052μs 58.4617 KOps/s 58.2217 KOps/s $\color{#35bf28}+0.41\%$
test_values[generalized_advantage_estimate-True-True] 10.0503ms 9.6834ms 103.2700 Ops/s 101.4604 Ops/s $\color{#35bf28}+1.78\%$
test_values[vec_generalized_advantage_estimate-True-True] 38.6815ms 33.5720ms 29.7867 Ops/s 29.6879 Ops/s $\color{#35bf28}+0.33\%$
test_values[td0_return_estimate-False-False] 0.2494ms 0.1901ms 5.2598 KOps/s 5.0251 KOps/s $\color{#35bf28}+4.67\%$
test_values[td1_return_estimate-False-False] 26.1280ms 24.3072ms 41.1401 Ops/s 40.7782 Ops/s $\color{#35bf28}+0.89\%$
test_values[vec_td1_return_estimate-False-False] 36.7591ms 33.7384ms 29.6398 Ops/s 29.6517 Ops/s $\color{#d91a1a}-0.04\%$
test_values[td_lambda_return_estimate-True-False] 35.2372ms 34.7008ms 28.8178 Ops/s 28.5193 Ops/s $\color{#35bf28}+1.05\%$
test_values[vec_td_lambda_return_estimate-True-False] 36.3883ms 33.7144ms 29.6609 Ops/s 29.6561 Ops/s $\color{#35bf28}+0.02\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 12.8378ms 8.4236ms 118.7144 Ops/s 118.8774 Ops/s $\color{#d91a1a}-0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.2300ms 1.8802ms 531.8658 Ops/s 533.0869 Ops/s $\color{#d91a1a}-0.23\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4579ms 0.3608ms 2.7720 KOps/s 2.7753 KOps/s $\color{#d91a1a}-0.12\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 46.8738ms 42.6172ms 23.4647 Ops/s 23.8174 Ops/s $\color{#d91a1a}-1.48\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.1123ms 3.0516ms 327.7001 Ops/s 326.0700 Ops/s $\color{#35bf28}+0.50\%$
test_dqn_speed[False-None] 7.0570ms 1.3997ms 714.4453 Ops/s 704.4057 Ops/s $\color{#35bf28}+1.43\%$
test_dqn_speed[False-backward] 1.9271ms 1.8682ms 535.2738 Ops/s 517.4340 Ops/s $\color{#35bf28}+3.45\%$
test_dqn_speed[True-None] 0.7296ms 0.4987ms 2.0051 KOps/s 2.0376 KOps/s $\color{#d91a1a}-1.59\%$
test_dqn_speed[True-backward] 0.9373ms 0.8975ms 1.1142 KOps/s 943.5726 Ops/s $\textbf{\color{#35bf28}+18.08\%}$
test_dqn_speed[reduce-overhead-None] 0.8477ms 0.4881ms 2.0490 KOps/s 2.0454 KOps/s $\color{#35bf28}+0.18\%$
test_dqn_speed[reduce-overhead-backward] 0.9351ms 0.8992ms 1.1122 KOps/s 1.0720 KOps/s $\color{#35bf28}+3.74\%$
test_ddpg_speed[False-None] 3.6172ms 2.9016ms 344.6379 Ops/s 339.4132 Ops/s $\color{#35bf28}+1.54\%$
test_ddpg_speed[False-backward] 5.3467ms 4.1577ms 240.5168 Ops/s 244.9379 Ops/s $\color{#d91a1a}-1.80\%$
test_ddpg_speed[True-None] 1.4594ms 1.0261ms 974.5993 Ops/s 974.1897 Ops/s $\color{#35bf28}+0.04\%$
test_ddpg_speed[True-backward] 2.0720ms 1.9461ms 513.8497 Ops/s 434.8795 Ops/s $\textbf{\color{#35bf28}+18.16\%}$
test_ddpg_speed[reduce-overhead-None] 1.5451ms 1.0352ms 965.9942 Ops/s 938.8631 Ops/s $\color{#35bf28}+2.89\%$
test_ddpg_speed[reduce-overhead-backward] 1.9661ms 1.9039ms 525.2445 Ops/s 510.5513 Ops/s $\color{#35bf28}+2.88\%$
test_sac_speed[False-None] 9.1848ms 8.0273ms 124.5750 Ops/s 120.5753 Ops/s $\color{#35bf28}+3.32\%$
test_sac_speed[False-backward] 11.6480ms 10.8128ms 92.4834 Ops/s 90.1856 Ops/s $\color{#35bf28}+2.55\%$
test_sac_speed[True-None] 2.3811ms 1.8657ms 535.9784 Ops/s 536.7799 Ops/s $\color{#d91a1a}-0.15\%$
test_sac_speed[True-backward] 3.8151ms 3.6452ms 274.3349 Ops/s 268.4614 Ops/s $\color{#35bf28}+2.19\%$
test_sac_speed[reduce-overhead-None] 2.5929ms 1.8675ms 535.4661 Ops/s 534.9049 Ops/s $\color{#35bf28}+0.10\%$
test_sac_speed[reduce-overhead-backward] 3.8180ms 3.5670ms 280.3445 Ops/s 276.3974 Ops/s $\color{#35bf28}+1.43\%$
test_redq_speed[False-None] 15.4549ms 13.2078ms 75.7126 Ops/s 75.3519 Ops/s $\color{#35bf28}+0.48\%$
test_redq_speed[False-backward] 24.1482ms 22.4754ms 44.4930 Ops/s 44.3044 Ops/s $\color{#35bf28}+0.43\%$
test_redq_speed[True-None] 6.1485ms 5.0982ms 196.1471 Ops/s 186.5825 Ops/s $\textbf{\color{#35bf28}+5.13\%}$
test_redq_speed[True-backward] 15.3760ms 12.9258ms 77.3646 Ops/s 77.8311 Ops/s $\color{#d91a1a}-0.60\%$
test_redq_speed[reduce-overhead-None] 6.7373ms 5.0621ms 197.5451 Ops/s 180.0433 Ops/s $\textbf{\color{#35bf28}+9.72\%}$
test_redq_speed[reduce-overhead-backward] 13.8168ms 12.8366ms 77.9024 Ops/s 76.3047 Ops/s $\color{#35bf28}+2.09\%$
test_redq_deprec_speed[False-None] 14.5848ms 13.1831ms 75.8547 Ops/s 71.6415 Ops/s $\textbf{\color{#35bf28}+5.88\%}$
test_redq_deprec_speed[False-backward] 19.3354ms 18.7995ms 53.1930 Ops/s 49.8337 Ops/s $\textbf{\color{#35bf28}+6.74\%}$
test_redq_deprec_speed[True-None] 4.1688ms 3.7330ms 267.8831 Ops/s 240.6817 Ops/s $\textbf{\color{#35bf28}+11.30\%}$
test_redq_deprec_speed[True-backward] 9.0141ms 8.5678ms 116.7160 Ops/s 120.4593 Ops/s $\color{#d91a1a}-3.11\%$
test_redq_deprec_speed[reduce-overhead-None] 4.8387ms 3.8578ms 259.2166 Ops/s 258.1993 Ops/s $\color{#35bf28}+0.39\%$
test_redq_deprec_speed[reduce-overhead-backward] 9.8298ms 8.9398ms 111.8590 Ops/s 122.0951 Ops/s $\textbf{\color{#d91a1a}-8.38\%}$
test_td3_speed[False-None] 34.9157ms 8.3140ms 120.2791 Ops/s 122.8850 Ops/s $\color{#d91a1a}-2.12\%$
test_td3_speed[False-backward] 10.8322ms 10.4048ms 96.1094 Ops/s 94.4808 Ops/s $\color{#35bf28}+1.72\%$
test_td3_speed[True-None] 1.8797ms 1.7396ms 574.8355 Ops/s 562.6969 Ops/s $\color{#35bf28}+2.16\%$
test_td3_speed[True-backward] 3.6327ms 3.4113ms 293.1439 Ops/s 282.2547 Ops/s $\color{#35bf28}+3.86\%$
test_td3_speed[reduce-overhead-None] 1.8770ms 1.7167ms 582.5063 Ops/s 562.7193 Ops/s $\color{#35bf28}+3.52\%$
test_td3_speed[reduce-overhead-backward] 3.6665ms 3.4215ms 292.2657 Ops/s 288.1401 Ops/s $\color{#35bf28}+1.43\%$
test_cql_speed[False-None] 38.5244ms 36.7693ms 27.1966 Ops/s 26.8258 Ops/s $\color{#35bf28}+1.38\%$
test_cql_speed[False-backward] 52.8524ms 47.7775ms 20.9304 Ops/s 20.8917 Ops/s $\color{#35bf28}+0.18\%$
test_cql_speed[True-None] 17.5185ms 15.7099ms 63.6540 Ops/s 62.0695 Ops/s $\color{#35bf28}+2.55\%$
test_cql_speed[True-backward] 24.0052ms 22.7405ms 43.9743 Ops/s 42.9059 Ops/s $\color{#35bf28}+2.49\%$
test_cql_speed[reduce-overhead-None] 17.1127ms 15.8498ms 63.0922 Ops/s 60.6102 Ops/s $\color{#35bf28}+4.09\%$
test_cql_speed[reduce-overhead-backward] 23.3523ms 22.1864ms 45.0726 Ops/s 43.1916 Ops/s $\color{#35bf28}+4.35\%$
test_a2c_speed[False-None] 7.9760ms 7.1040ms 140.7662 Ops/s 131.8811 Ops/s $\textbf{\color{#35bf28}+6.74\%}$
test_a2c_speed[False-backward] 14.5827ms 14.2114ms 70.3660 Ops/s 66.3384 Ops/s $\textbf{\color{#35bf28}+6.07\%}$
test_a2c_speed[True-None] 5.0061ms 4.1987ms 238.1680 Ops/s 233.6929 Ops/s $\color{#35bf28}+1.91\%$
test_a2c_speed[True-backward] 11.6044ms 10.6810ms 93.6240 Ops/s 88.3891 Ops/s $\textbf{\color{#35bf28}+5.92\%}$
test_a2c_speed[reduce-overhead-None] 4.9318ms 4.1979ms 238.2146 Ops/s 230.8473 Ops/s $\color{#35bf28}+3.19\%$
test_a2c_speed[reduce-overhead-backward] 11.2904ms 10.9085ms 91.6718 Ops/s 88.5731 Ops/s $\color{#35bf28}+3.50\%$
test_ppo_speed[False-None] 10.0012ms 7.4459ms 134.3026 Ops/s 128.7653 Ops/s $\color{#35bf28}+4.30\%$
test_ppo_speed[False-backward] 16.2581ms 14.7360ms 67.8612 Ops/s 65.2865 Ops/s $\color{#35bf28}+3.94\%$
test_ppo_speed[True-None] 4.1764ms 3.6824ms 271.5603 Ops/s 264.0152 Ops/s $\color{#35bf28}+2.86\%$
test_ppo_speed[True-backward] 10.1764ms 9.6093ms 104.0660 Ops/s 99.5112 Ops/s $\color{#35bf28}+4.58\%$
test_ppo_speed[reduce-overhead-None] 3.9795ms 3.6986ms 270.3730 Ops/s 264.9999 Ops/s $\color{#35bf28}+2.03\%$
test_ppo_speed[reduce-overhead-backward] 10.8928ms 9.5571ms 104.6342 Ops/s 97.1253 Ops/s $\textbf{\color{#35bf28}+7.73\%}$
test_reinforce_speed[False-None] 7.4997ms 6.5380ms 152.9521 Ops/s 144.8956 Ops/s $\textbf{\color{#35bf28}+5.56\%}$
test_reinforce_speed[False-backward] 10.2383ms 9.8035ms 102.0046 Ops/s 96.7893 Ops/s $\textbf{\color{#35bf28}+5.39\%}$
test_reinforce_speed[True-None] 3.3114ms 2.6869ms 372.1783 Ops/s 368.6518 Ops/s $\color{#35bf28}+0.96\%$
test_reinforce_speed[True-backward] 9.5857ms 8.8445ms 113.0647 Ops/s 111.5334 Ops/s $\color{#35bf28}+1.37\%$
test_reinforce_speed[reduce-overhead-None] 3.0267ms 2.6524ms 377.0201 Ops/s 294.6025 Ops/s $\textbf{\color{#35bf28}+27.98\%}$
test_reinforce_speed[reduce-overhead-backward] 8.9527ms 8.5359ms 117.1518 Ops/s 110.5512 Ops/s $\textbf{\color{#35bf28}+5.97\%}$
test_iql_speed[False-None] 33.9537ms 32.1054ms 31.1474 Ops/s 29.7602 Ops/s $\color{#35bf28}+4.66\%$
test_iql_speed[False-backward] 54.3540ms 45.3574ms 22.0471 Ops/s 21.3481 Ops/s $\color{#35bf28}+3.27\%$
test_iql_speed[True-None] 12.0631ms 10.6657ms 93.7588 Ops/s 89.0193 Ops/s $\textbf{\color{#35bf28}+5.32\%}$
test_iql_speed[True-backward] 22.1125ms 21.3187ms 46.9071 Ops/s 44.1747 Ops/s $\textbf{\color{#35bf28}+6.19\%}$
test_iql_speed[reduce-overhead-None] 11.8580ms 10.7855ms 92.7171 Ops/s 89.7226 Ops/s $\color{#35bf28}+3.34\%$
test_iql_speed[reduce-overhead-backward] 22.1840ms 21.4908ms 46.5316 Ops/s 44.1988 Ops/s $\textbf{\color{#35bf28}+5.28\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5580ms 4.8036ms 208.1767 Ops/s 193.0198 Ops/s $\textbf{\color{#35bf28}+7.85\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8938ms 0.5137ms 1.9468 KOps/s 1.8755 KOps/s $\color{#35bf28}+3.80\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8232ms 0.4885ms 2.0472 KOps/s 1.9993 KOps/s $\color{#35bf28}+2.40\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 4.8980ms 4.5996ms 217.4100 Ops/s 206.8375 Ops/s $\textbf{\color{#35bf28}+5.11\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.7846ms 0.5022ms 1.9911 KOps/s 1.9385 KOps/s $\color{#35bf28}+2.71\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8181ms 0.4881ms 2.0489 KOps/s 2.0451 KOps/s $\color{#35bf28}+0.18\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.0900ms 1.6257ms 615.1299 Ops/s 584.9417 Ops/s $\textbf{\color{#35bf28}+5.16\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.9408ms 1.5770ms 634.1289 Ops/s 620.8704 Ops/s $\color{#35bf28}+2.14\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.7329ms 4.7613ms 210.0248 Ops/s 200.3914 Ops/s $\color{#35bf28}+4.81\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0826ms 0.6376ms 1.5684 KOps/s 1.5101 KOps/s $\color{#35bf28}+3.86\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7921ms 0.6107ms 1.6376 KOps/s 1.5815 KOps/s $\color{#35bf28}+3.55\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 4.8724ms 4.6277ms 216.0901 Ops/s 207.9770 Ops/s $\color{#35bf28}+3.90\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1682ms 0.5160ms 1.9381 KOps/s 1.8660 KOps/s $\color{#35bf28}+3.86\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7113ms 0.4879ms 2.0497 KOps/s 1.9820 KOps/s $\color{#35bf28}+3.42\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.0275ms 4.6215ms 216.3778 Ops/s 209.2084 Ops/s $\color{#35bf28}+3.43\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0001ms 0.4983ms 2.0067 KOps/s 1.9645 KOps/s $\color{#35bf28}+2.15\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8178ms 0.4853ms 2.0607 KOps/s 1.9936 KOps/s $\color{#35bf28}+3.37\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.6337ms 4.6946ms 213.0128 Ops/s 202.2327 Ops/s $\textbf{\color{#35bf28}+5.33\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8036ms 0.6357ms 1.5730 KOps/s 1.5222 KOps/s $\color{#35bf28}+3.34\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9867ms 0.6172ms 1.6202 KOps/s 1.5683 KOps/s $\color{#35bf28}+3.31\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.3847s 11.8152ms 84.6364 Ops/s 37.4604 Ops/s $\textbf{\color{#35bf28}+125.94\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.9994ms 2.2258ms 449.2838 Ops/s 435.6363 Ops/s $\color{#35bf28}+3.13\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4806ms 1.4323ms 698.1747 Ops/s 748.3704 Ops/s $\textbf{\color{#d91a1a}-6.71\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.1609ms 4.1876ms 238.8019 Ops/s 227.4651 Ops/s $\color{#35bf28}+4.98\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.2742ms 2.3080ms 433.2715 Ops/s 395.0608 Ops/s $\textbf{\color{#35bf28}+9.67\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.8669ms 1.3133ms 761.4359 Ops/s 831.9422 Ops/s $\textbf{\color{#d91a1a}-8.47\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.3601s 11.5562ms 86.5338 Ops/s 227.3207 Ops/s $\textbf{\color{#d91a1a}-61.93\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.1482ms 2.4081ms 415.2611 Ops/s 398.3587 Ops/s $\color{#35bf28}+4.24\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 5.0606ms 1.4442ms 692.4481 Ops/s 706.1411 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 12.9718ms 12.7907ms 78.1815 Ops/s 72.5251 Ops/s $\textbf{\color{#35bf28}+7.80\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.5677ms 15.0093ms 66.6254 Ops/s 66.2966 Ops/s $\color{#35bf28}+0.50\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 23.0754ms 21.4282ms 46.6675 Ops/s 44.1260 Ops/s $\textbf{\color{#35bf28}+5.76\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.5360ms 14.9591ms 66.8489 Ops/s 66.2182 Ops/s $\color{#35bf28}+0.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.9148ms 21.3856ms 46.7604 Ops/s 44.8244 Ops/s $\color{#35bf28}+4.32\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 18.2704ms 16.3745ms 61.0705 Ops/s 60.6823 Ops/s $\color{#35bf28}+0.64\%$

Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}17$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7233s 0.7203s 1.3884 Ops/s 1.3557 Ops/s $\color{#35bf28}+2.41\%$
test_transformed 0.9680s 0.9660s 1.0352 Ops/s 1.0374 Ops/s $\color{#d91a1a}-0.22\%$
test_serial 2.2338s 2.1535s 0.4644 Ops/s 0.4696 Ops/s $\color{#d91a1a}-1.11\%$
test_parallel 2.0130s 1.9583s 0.5106 Ops/s 0.5013 Ops/s $\color{#35bf28}+1.86\%$
test_step_mdp_speed[True-True-True-True-True] 0.2203ms 40.9035μs 24.4478 KOps/s 24.7128 KOps/s $\color{#d91a1a}-1.07\%$
test_step_mdp_speed[True-True-True-True-False] 55.6110μs 23.5522μs 42.4588 KOps/s 43.0161 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[True-True-True-False-True] 46.4410μs 23.0014μs 43.4756 KOps/s 45.9662 KOps/s $\textbf{\color{#d91a1a}-5.42\%}$
test_step_mdp_speed[True-True-True-False-False] 39.8210μs 13.2387μs 75.5361 KOps/s 76.8591 KOps/s $\color{#d91a1a}-1.72\%$
test_step_mdp_speed[True-True-False-True-True] 75.4110μs 43.5023μs 22.9873 KOps/s 23.7956 KOps/s $\color{#d91a1a}-3.40\%$
test_step_mdp_speed[True-True-False-True-False] 53.2310μs 26.1106μs 38.2986 KOps/s 39.0294 KOps/s $\color{#d91a1a}-1.87\%$
test_step_mdp_speed[True-True-False-False-True] 51.2510μs 25.2242μs 39.6445 KOps/s 40.5359 KOps/s $\color{#d91a1a}-2.20\%$
test_step_mdp_speed[True-True-False-False-False] 43.6110μs 15.5481μs 64.3167 KOps/s 65.1869 KOps/s $\color{#d91a1a}-1.34\%$
test_step_mdp_speed[True-False-True-True-True] 80.3810μs 46.1110μs 21.6868 KOps/s 22.6855 KOps/s $\color{#d91a1a}-4.40\%$
test_step_mdp_speed[True-False-True-True-False] 65.2620μs 28.0898μs 35.6001 KOps/s 35.5823 KOps/s $\color{#35bf28}+0.05\%$
test_step_mdp_speed[True-False-True-False-True] 50.8210μs 25.3725μs 39.4127 KOps/s 41.2004 KOps/s $\color{#d91a1a}-4.34\%$
test_step_mdp_speed[True-False-True-False-False] 44.4500μs 15.3530μs 65.1338 KOps/s 65.7392 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[True-False-False-True-True] 74.5820μs 48.6002μs 20.5760 KOps/s 21.5419 KOps/s $\color{#d91a1a}-4.48\%$
test_step_mdp_speed[True-False-False-True-False] 57.5510μs 30.8528μs 32.4120 KOps/s 33.0820 KOps/s $\color{#d91a1a}-2.03\%$
test_step_mdp_speed[True-False-False-False-True] 52.8010μs 27.3317μs 36.5875 KOps/s 36.8009 KOps/s $\color{#d91a1a}-0.58\%$
test_step_mdp_speed[True-False-False-False-False] 42.4010μs 17.6364μs 56.7010 KOps/s 56.8448 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[False-True-True-True-True] 85.9720μs 46.2413μs 21.6257 KOps/s 22.2315 KOps/s $\color{#d91a1a}-2.73\%$
test_step_mdp_speed[False-True-True-True-False] 55.3810μs 28.4421μs 35.1592 KOps/s 35.6270 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[False-True-True-False-True] 70.4810μs 29.6335μs 33.7455 KOps/s 35.4951 KOps/s $\color{#d91a1a}-4.93\%$
test_step_mdp_speed[False-True-True-False-False] 42.1910μs 17.4671μs 57.2506 KOps/s 59.2367 KOps/s $\color{#d91a1a}-3.35\%$
test_step_mdp_speed[False-True-False-True-True] 77.7320μs 47.6570μs 20.9833 KOps/s 21.7742 KOps/s $\color{#d91a1a}-3.63\%$
test_step_mdp_speed[False-True-False-True-False] 62.2510μs 30.5511μs 32.7321 KOps/s 33.5912 KOps/s $\color{#d91a1a}-2.56\%$
test_step_mdp_speed[False-True-False-False-True] 3.3102ms 32.2421μs 31.0153 KOps/s 32.5979 KOps/s $\color{#d91a1a}-4.85\%$
test_step_mdp_speed[False-True-False-False-False] 41.5610μs 19.8544μs 50.3666 KOps/s 52.2226 KOps/s $\color{#d91a1a}-3.55\%$
test_step_mdp_speed[False-False-True-True-True] 82.4520μs 50.3811μs 19.8487 KOps/s 20.8778 KOps/s $\color{#d91a1a}-4.93\%$
test_step_mdp_speed[False-False-True-True-False] 58.6520μs 32.9929μs 30.3096 KOps/s 31.0404 KOps/s $\color{#d91a1a}-2.35\%$
test_step_mdp_speed[False-False-True-False-True] 60.9410μs 31.8114μs 31.4353 KOps/s 32.9181 KOps/s $\color{#d91a1a}-4.50\%$
test_step_mdp_speed[False-False-True-False-False] 48.3310μs 19.4516μs 51.4098 KOps/s 52.1841 KOps/s $\color{#d91a1a}-1.48\%$
test_step_mdp_speed[False-False-False-True-True] 83.8420μs 52.4210μs 19.0763 KOps/s 19.6471 KOps/s $\color{#d91a1a}-2.91\%$
test_step_mdp_speed[False-False-False-True-False] 69.4410μs 35.5771μs 28.1080 KOps/s 28.4908 KOps/s $\color{#d91a1a}-1.34\%$
test_step_mdp_speed[False-False-False-False-True] 61.0310μs 33.2023μs 30.1184 KOps/s 31.0164 KOps/s $\color{#d91a1a}-2.90\%$
test_step_mdp_speed[False-False-False-False-False] 46.9410μs 21.9702μs 45.5163 KOps/s 47.8968 KOps/s $\color{#d91a1a}-4.97\%$
test_values[generalized_advantage_estimate-True-True] 25.3003ms 24.9359ms 40.1029 Ops/s 39.6996 Ops/s $\color{#35bf28}+1.02\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1008s 2.9209ms 342.3625 Ops/s 352.3345 Ops/s $\color{#d91a1a}-2.83\%$
test_values[td0_return_estimate-False-False] 0.1105ms 80.5333μs 12.4172 KOps/s 12.3336 KOps/s $\color{#35bf28}+0.68\%$
test_values[td1_return_estimate-False-False] 58.8318ms 56.6316ms 17.6580 Ops/s 18.0686 Ops/s $\color{#d91a1a}-2.27\%$
test_values[vec_td1_return_estimate-False-False] 1.3742ms 1.0905ms 916.9963 Ops/s 917.8099 Ops/s $\color{#d91a1a}-0.09\%$
test_values[td_lambda_return_estimate-True-False] 93.1800ms 87.7618ms 11.3945 Ops/s 11.3934 Ops/s $+0.01\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3586ms 1.0862ms 920.6765 Ops/s 916.6481 Ops/s $\color{#35bf28}+0.44\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.1937ms 25.7270ms 38.8696 Ops/s 40.1564 Ops/s $\color{#d91a1a}-3.20\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0452ms 0.7605ms 1.3150 KOps/s 1.3049 KOps/s $\color{#35bf28}+0.77\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7669ms 0.6790ms 1.4728 KOps/s 1.4643 KOps/s $\color{#35bf28}+0.58\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5160ms 1.4815ms 675.0096 Ops/s 672.4408 Ops/s $\color{#35bf28}+0.38\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7353ms 0.6946ms 1.4396 KOps/s 1.4381 KOps/s $\color{#35bf28}+0.10\%$
test_dqn_speed[False-None] 7.0312ms 1.5382ms 650.1199 Ops/s 653.9915 Ops/s $\color{#d91a1a}-0.59\%$
test_dqn_speed[False-backward] 2.2852ms 2.1617ms 462.6073 Ops/s 469.5299 Ops/s $\color{#d91a1a}-1.47\%$
test_dqn_speed[True-None] 1.0563ms 0.5753ms 1.7383 KOps/s 1.7895 KOps/s $\color{#d91a1a}-2.86\%$
test_dqn_speed[True-backward] 1.2346ms 1.1292ms 885.5893 Ops/s 864.2971 Ops/s $\color{#35bf28}+2.46\%$
test_dqn_speed[reduce-overhead-None] 0.6673ms 0.5718ms 1.7487 KOps/s 1.7320 KOps/s $\color{#35bf28}+0.97\%$
test_dqn_speed[reduce-overhead-backward] 1.0467ms 0.9879ms 1.0122 KOps/s 992.7814 Ops/s $\color{#35bf28}+1.96\%$
test_ddpg_speed[False-None] 3.2295ms 2.9022ms 344.5688 Ops/s 343.8146 Ops/s $\color{#35bf28}+0.22\%$
test_ddpg_speed[False-backward] 4.4737ms 4.1767ms 239.4223 Ops/s 241.0217 Ops/s $\color{#d91a1a}-0.66\%$
test_ddpg_speed[True-None] 1.2653ms 1.0961ms 912.3401 Ops/s 894.2927 Ops/s $\color{#35bf28}+2.02\%$
test_ddpg_speed[True-backward] 2.2602ms 2.1940ms 455.7900 Ops/s 422.8237 Ops/s $\textbf{\color{#35bf28}+7.80\%}$
test_ddpg_speed[reduce-overhead-None] 1.6031ms 1.1093ms 901.5009 Ops/s 887.6222 Ops/s $\color{#35bf28}+1.56\%$
test_ddpg_speed[reduce-overhead-backward] 1.8128ms 1.6934ms 590.5193 Ops/s 545.3052 Ops/s $\textbf{\color{#35bf28}+8.29\%}$
test_sac_speed[False-None] 9.4312ms 8.1056ms 123.3713 Ops/s 123.2063 Ops/s $\color{#35bf28}+0.13\%$
test_sac_speed[False-backward] 11.8235ms 11.3771ms 87.8962 Ops/s 88.4099 Ops/s $\color{#d91a1a}-0.58\%$
test_sac_speed[True-None] 1.9523ms 1.5723ms 636.0226 Ops/s 636.9764 Ops/s $\color{#d91a1a}-0.15\%$
test_sac_speed[True-backward] 3.4904ms 3.3351ms 299.8377 Ops/s 285.8139 Ops/s $\color{#35bf28}+4.91\%$
test_sac_speed[reduce-overhead-None] 22.8230ms 12.4495ms 80.3247 Ops/s 81.3335 Ops/s $\color{#d91a1a}-1.24\%$
test_sac_speed[reduce-overhead-backward] 1.6504ms 1.5412ms 648.8495 Ops/s 649.0121 Ops/s $\color{#d91a1a}-0.03\%$
test_redq_speed[False-None] 8.4136ms 7.5826ms 131.8805 Ops/s 130.6522 Ops/s $\color{#35bf28}+0.94\%$
test_redq_speed[False-backward] 12.5336ms 11.7733ms 84.9378 Ops/s 85.0508 Ops/s $\color{#d91a1a}-0.13\%$
test_redq_speed[True-None] 2.1183ms 2.0277ms 493.1766 Ops/s 494.2291 Ops/s $\color{#d91a1a}-0.21\%$
test_redq_speed[True-backward] 3.8025ms 3.7324ms 267.9240 Ops/s 252.9315 Ops/s $\textbf{\color{#35bf28}+5.93\%}$
test_redq_speed[reduce-overhead-None] 2.4399ms 2.0287ms 492.9165 Ops/s 480.7739 Ops/s $\color{#35bf28}+2.53\%$
test_redq_speed[reduce-overhead-backward] 4.3624ms 3.9139ms 255.5023 Ops/s 270.5126 Ops/s $\textbf{\color{#d91a1a}-5.55\%}$
test_redq_deprec_speed[False-None] 9.7604ms 9.2117ms 108.5579 Ops/s 107.9726 Ops/s $\color{#35bf28}+0.54\%$
test_redq_deprec_speed[False-backward] 13.2253ms 12.4758ms 80.1549 Ops/s 81.3028 Ops/s $\color{#d91a1a}-1.41\%$
test_redq_deprec_speed[True-None] 2.8347ms 2.3830ms 419.6315 Ops/s 398.6446 Ops/s $\textbf{\color{#35bf28}+5.26\%}$
test_redq_deprec_speed[True-backward] 4.3023ms 4.2356ms 236.0929 Ops/s 244.4708 Ops/s $\color{#d91a1a}-3.43\%$
test_redq_deprec_speed[reduce-overhead-None] 2.8274ms 2.3712ms 421.7259 Ops/s 421.8509 Ops/s $\color{#d91a1a}-0.03\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.4870ms 4.3810ms 228.2588 Ops/s 234.4532 Ops/s $\color{#d91a1a}-2.64\%$
test_td3_speed[False-None] 8.0981ms 8.0601ms 124.0679 Ops/s 125.7868 Ops/s $\color{#d91a1a}-1.37\%$
test_td3_speed[False-backward] 11.2956ms 10.6647ms 93.7672 Ops/s 95.0524 Ops/s $\color{#d91a1a}-1.35\%$
test_td3_speed[True-None] 1.7343ms 1.6793ms 595.4770 Ops/s 615.5979 Ops/s $\color{#d91a1a}-3.27\%$
test_td3_speed[True-backward] 3.6602ms 3.1936ms 313.1221 Ops/s 298.5414 Ops/s $\color{#35bf28}+4.88\%$
test_td3_speed[reduce-overhead-None] 80.6819ms 26.1020ms 38.3113 Ops/s 37.0803 Ops/s $\color{#35bf28}+3.32\%$
test_td3_speed[reduce-overhead-backward] 1.4490ms 1.3406ms 745.9102 Ops/s 741.4678 Ops/s $\color{#35bf28}+0.60\%$
test_cql_speed[False-None] 17.6452ms 17.0466ms 58.6628 Ops/s 58.1589 Ops/s $\color{#35bf28}+0.87\%$
test_cql_speed[False-backward] 22.8044ms 22.2615ms 44.9205 Ops/s 45.2853 Ops/s $\color{#d91a1a}-0.81\%$
test_cql_speed[True-None] 3.4412ms 2.9970ms 333.6710 Ops/s 333.8269 Ops/s $\color{#d91a1a}-0.05\%$
test_cql_speed[True-backward] 5.6379ms 5.1941ms 192.5249 Ops/s 191.4467 Ops/s $\color{#35bf28}+0.56\%$
test_cql_speed[reduce-overhead-None] 21.3672ms 13.1255ms 76.1878 Ops/s 76.6399 Ops/s $\color{#d91a1a}-0.59\%$
test_cql_speed[reduce-overhead-backward] 1.7733ms 1.7281ms 578.6545 Ops/s 578.1400 Ops/s $\color{#35bf28}+0.09\%$
test_a2c_speed[False-None] 3.4424ms 3.3103ms 302.0894 Ops/s 301.4277 Ops/s $\color{#35bf28}+0.22\%$
test_a2c_speed[False-backward] 6.8979ms 6.4340ms 155.4245 Ops/s 155.5028 Ops/s $\color{#d91a1a}-0.05\%$
test_a2c_speed[True-None] 1.1095ms 1.0280ms 972.7419 Ops/s 976.5786 Ops/s $\color{#d91a1a}-0.39\%$
test_a2c_speed[True-backward] 2.6845ms 2.6456ms 377.9855 Ops/s 353.2496 Ops/s $\textbf{\color{#35bf28}+7.00\%}$
test_a2c_speed[reduce-overhead-None] 20.7798ms 11.2548ms 88.8511 Ops/s 87.5394 Ops/s $\color{#35bf28}+1.50\%$
test_a2c_speed[reduce-overhead-backward] 1.0434ms 0.9956ms 1.0045 KOps/s 854.6348 Ops/s $\textbf{\color{#35bf28}+17.53\%}$
test_ppo_speed[False-None] 3.9832ms 3.7424ms 267.2058 Ops/s 268.7270 Ops/s $\color{#d91a1a}-0.57\%$
test_ppo_speed[False-backward] 7.2919ms 6.8553ms 145.8718 Ops/s 141.2490 Ops/s $\color{#35bf28}+3.27\%$
test_ppo_speed[True-None] 1.0430ms 0.9748ms 1.0258 KOps/s 1.0389 KOps/s $\color{#d91a1a}-1.25\%$
test_ppo_speed[True-backward] 2.6525ms 2.5925ms 385.7217 Ops/s 359.6970 Ops/s $\textbf{\color{#35bf28}+7.24\%}$
test_ppo_speed[reduce-overhead-None] 0.6724ms 0.5336ms 1.8739 KOps/s 1.8410 KOps/s $\color{#35bf28}+1.79\%$
test_ppo_speed[reduce-overhead-backward] 1.0559ms 0.9934ms 1.0066 KOps/s 880.8368 Ops/s $\textbf{\color{#35bf28}+14.28\%}$
test_reinforce_speed[False-None] 2.4438ms 2.3023ms 434.3459 Ops/s 436.2873 Ops/s $\color{#d91a1a}-0.44\%$
test_reinforce_speed[False-backward] 3.7796ms 3.3250ms 300.7512 Ops/s 292.3651 Ops/s $\color{#35bf28}+2.87\%$
test_reinforce_speed[True-None] 0.9105ms 0.8481ms 1.1791 KOps/s 1.1643 KOps/s $\color{#35bf28}+1.27\%$
test_reinforce_speed[True-backward] 2.5218ms 2.4573ms 406.9549 Ops/s 379.8340 Ops/s $\textbf{\color{#35bf28}+7.14\%}$
test_reinforce_speed[reduce-overhead-None] 22.0491ms 11.5955ms 86.2406 Ops/s 87.7962 Ops/s $\color{#d91a1a}-1.77\%$
test_reinforce_speed[reduce-overhead-backward] 1.0957ms 1.0565ms 946.5230 Ops/s 811.8187 Ops/s $\textbf{\color{#35bf28}+16.59\%}$
test_iql_speed[False-None] 9.9134ms 9.4355ms 105.9828 Ops/s 106.7096 Ops/s $\color{#d91a1a}-0.68\%$
test_iql_speed[False-backward] 13.5822ms 13.1690ms 75.9359 Ops/s 74.4701 Ops/s $\color{#35bf28}+1.97\%$
test_iql_speed[True-None] 1.9771ms 1.7965ms 556.6498 Ops/s 565.6017 Ops/s $\color{#d91a1a}-1.58\%$
test_iql_speed[True-backward] 4.3528ms 4.2950ms 232.8315 Ops/s 221.9190 Ops/s $\color{#35bf28}+4.92\%$
test_iql_speed[reduce-overhead-None] 19.9207ms 11.4848ms 87.0717 Ops/s 88.2491 Ops/s $\color{#d91a1a}-1.33\%$
test_iql_speed[reduce-overhead-backward] 1.5202ms 1.4634ms 683.3463 Ops/s 619.3002 Ops/s $\textbf{\color{#35bf28}+10.34\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9748ms 6.3677ms 157.0436 Ops/s 152.3090 Ops/s $\color{#35bf28}+3.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6397ms 0.3629ms 2.7556 KOps/s 3.3148 KOps/s $\textbf{\color{#d91a1a}-16.87\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5818ms 0.3416ms 2.9278 KOps/s 3.8348 KOps/s $\textbf{\color{#d91a1a}-23.65\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3800ms 6.1498ms 162.6075 Ops/s 159.9831 Ops/s $\color{#35bf28}+1.64\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9913ms 0.3001ms 3.3323 KOps/s 3.7269 KOps/s $\textbf{\color{#d91a1a}-10.59\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5024ms 0.2894ms 3.4551 KOps/s 3.7583 KOps/s $\textbf{\color{#d91a1a}-8.07\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5419ms 1.3520ms 739.6551 Ops/s 786.8333 Ops/s $\textbf{\color{#d91a1a}-6.00\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4855ms 1.2997ms 769.3909 Ops/s 822.8109 Ops/s $\textbf{\color{#d91a1a}-6.49\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4011ms 6.3002ms 158.7254 Ops/s 156.5058 Ops/s $\color{#35bf28}+1.42\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1746ms 0.4523ms 2.2108 KOps/s 2.2634 KOps/s $\color{#d91a1a}-2.32\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7099ms 0.4579ms 2.1840 KOps/s 2.3021 KOps/s $\textbf{\color{#d91a1a}-5.13\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2628ms 6.1605ms 162.3249 Ops/s 161.2329 Ops/s $\color{#35bf28}+0.68\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.4489ms 0.3477ms 2.8759 KOps/s 3.6601 KOps/s $\textbf{\color{#d91a1a}-21.43\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5605ms 0.3596ms 2.7812 KOps/s 3.0749 KOps/s $\textbf{\color{#d91a1a}-9.55\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3534ms 6.1278ms 163.1899 Ops/s 161.0787 Ops/s $\color{#35bf28}+1.31\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9535ms 0.3009ms 3.3236 KOps/s 3.2403 KOps/s $\color{#35bf28}+2.57\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5521ms 0.3026ms 3.3049 KOps/s 2.9911 KOps/s $\textbf{\color{#35bf28}+10.49\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4834ms 6.3401ms 157.7265 Ops/s 156.8392 Ops/s $\color{#35bf28}+0.57\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9018ms 0.4718ms 2.1195 KOps/s 2.0388 KOps/s $\color{#35bf28}+3.95\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6848ms 0.4512ms 2.2164 KOps/s 2.3447 KOps/s $\textbf{\color{#d91a1a}-5.47\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1728ms 5.5018ms 181.7573 Ops/s 182.4799 Ops/s $\color{#d91a1a}-0.40\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.8115ms 1.9211ms 520.5314 Ops/s 438.4669 Ops/s $\textbf{\color{#35bf28}+18.72\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.8151ms 1.2131ms 824.3568 Ops/s 815.3634 Ops/s $\color{#35bf28}+1.10\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.4665ms 5.5764ms 179.3280 Ops/s 182.3049 Ops/s $\color{#d91a1a}-1.63\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.4784ms 2.0177ms 495.6123 Ops/s 445.3061 Ops/s $\textbf{\color{#35bf28}+11.30\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1961ms 1.2414ms 805.5509 Ops/s 762.2288 Ops/s $\textbf{\color{#35bf28}+5.68\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5112s 15.8893ms 62.9354 Ops/s 32.8454 Ops/s $\textbf{\color{#35bf28}+91.61\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.7708ms 2.2110ms 452.2827 Ops/s 436.8929 Ops/s $\color{#35bf28}+3.52\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.2062ms 1.3411ms 745.6478 Ops/s 698.1352 Ops/s $\textbf{\color{#35bf28}+6.81\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 16.2247ms 15.4881ms 64.5657 Ops/s 64.1330 Ops/s $\color{#35bf28}+0.67\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.0931ms 18.1537ms 55.0852 Ops/s 56.2418 Ops/s $\color{#d91a1a}-2.06\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 20.2956ms 19.8068ms 50.4877 Ops/s 48.3766 Ops/s $\color{#35bf28}+4.36\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.2640ms 18.2618ms 54.7590 Ops/s 56.0046 Ops/s $\color{#d91a1a}-2.22\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.4590ms 19.7722ms 50.5761 Ops/s 48.5998 Ops/s $\color{#35bf28}+4.07\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.5925ms 19.8194ms 50.4557 Ops/s 50.9672 Ops/s $\color{#d91a1a}-1.00\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants