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.

Sign up for GitHub

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

Jump to bottom

[Feature] spec.cardinality #2638

Merged
merged 4 commits into from
Dec 12, 2024
Merged

[Feature] spec.cardinality #2638

merged 4 commits into from
Dec 12, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Dec 6, 2024 •
edited
Loading

@vmoens
Update
84692bc 
[ghstack-poisoned]
@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Dec 6, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2638

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

❌ 8 New Failures, 1 Cancelled Job, 17 Unrelated Failures

As of commit b827784 with merge base 57dc25a (image):

NEW FAILURES - The following jobs have failed:

CANCELLED JOB - The following job was cancelled. Please retry:

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

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.

This was referenced Dec 6, 2024
Merged
Merged
Merged
Closed
Merged
Merged
Merged
@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 6, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Dec 6, 2024 •
edited
Loading

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4362s 0.4344s 2.3023 Ops/s 2.1906 Ops/s $\textbf{\color{#35bf28}+5.10\%}$
test_transformed 0.6033s 0.6016s 1.6623 Ops/s 1.5710 Ops/s $\textbf{\color{#35bf28}+5.81\%}$
test_serial 1.3573s 1.3457s 0.7431 Ops/s 0.7289 Ops/s $\color{#35bf28}+1.95\%$
test_parallel 1.2989s 1.2903s 0.7750 Ops/s 0.7563 Ops/s $\color{#35bf28}+2.47\%$
test_step_mdp_speed[True-True-True-True-True] 0.2131ms 30.9387μs 32.3219 KOps/s 33.0604 KOps/s $\color{#d91a1a}-2.23\%$
test_step_mdp_speed[True-True-True-True-False] 46.3570μs 18.5130μs 54.0161 KOps/s 56.9413 KOps/s $\textbf{\color{#d91a1a}-5.14\%}$
test_step_mdp_speed[True-True-True-False-True] 51.6260μs 17.6258μs 56.7350 KOps/s 59.0858 KOps/s $\color{#d91a1a}-3.98\%$
test_step_mdp_speed[True-True-True-False-False] 31.5990μs 10.5726μs 94.5837 KOps/s 101.4275 KOps/s $\textbf{\color{#d91a1a}-6.75\%}$
test_step_mdp_speed[True-True-False-True-True] 74.9090μs 32.8396μs 30.4511 KOps/s 30.9592 KOps/s $\color{#d91a1a}-1.64\%$
test_step_mdp_speed[True-True-False-True-False] 58.8700μs 20.5764μs 48.5994 KOps/s 51.0009 KOps/s $\color{#d91a1a}-4.71\%$
test_step_mdp_speed[True-True-False-False-True] 70.1210μs 19.4275μs 51.4734 KOps/s 53.3381 KOps/s $\color{#d91a1a}-3.50\%$
test_step_mdp_speed[True-True-False-False-False] 47.9190μs 12.4205μs 80.5122 KOps/s 84.4970 KOps/s $\color{#d91a1a}-4.72\%$
test_step_mdp_speed[True-False-True-True-True] 75.8210μs 34.8027μs 28.7334 KOps/s 29.3456 KOps/s $\color{#d91a1a}-2.09\%$
test_step_mdp_speed[True-False-True-True-False] 65.9430μs 22.2990μs 44.8450 KOps/s 46.4193 KOps/s $\color{#d91a1a}-3.39\%$
test_step_mdp_speed[True-False-True-False-True] 0.2066ms 20.0117μs 49.9707 KOps/s 53.5652 KOps/s $\textbf{\color{#d91a1a}-6.71\%}$
test_step_mdp_speed[True-False-True-False-False] 45.1740μs 12.3969μs 80.6651 KOps/s 84.9912 KOps/s $\textbf{\color{#d91a1a}-5.09\%}$
test_step_mdp_speed[True-False-False-True-True] 0.3104ms 37.4290μs 26.7173 KOps/s 27.8918 KOps/s $\color{#d91a1a}-4.21\%$
test_step_mdp_speed[True-False-False-True-False] 51.9170μs 24.2426μs 41.2498 KOps/s 43.1896 KOps/s $\color{#d91a1a}-4.49\%$
test_step_mdp_speed[True-False-False-False-True] 64.7200μs 21.3750μs 46.7836 KOps/s 48.8247 KOps/s $\color{#d91a1a}-4.18\%$
test_step_mdp_speed[True-False-False-False-False] 47.8190μs 14.2977μs 69.9415 KOps/s 72.9156 KOps/s $\color{#d91a1a}-4.08\%$
test_step_mdp_speed[False-True-True-True-True] 79.7980μs 34.9606μs 28.6036 KOps/s 29.0919 KOps/s $\color{#d91a1a}-1.68\%$
test_step_mdp_speed[False-True-True-True-False] 78.9480μs 22.5890μs 44.2693 KOps/s 45.6469 KOps/s $\color{#d91a1a}-3.02\%$
test_step_mdp_speed[False-True-True-False-True] 0.1914ms 22.2974μs 44.8482 KOps/s 46.3996 KOps/s $\color{#d91a1a}-3.34\%$
test_step_mdp_speed[False-True-True-False-False] 0.1395ms 13.8891μs 71.9988 KOps/s 77.9313 KOps/s $\textbf{\color{#d91a1a}-7.61\%}$
test_step_mdp_speed[False-True-False-True-True] 0.2236ms 36.6157μs 27.3107 KOps/s 28.4720 KOps/s $\color{#d91a1a}-4.08\%$
test_step_mdp_speed[False-True-False-True-False] 68.1870μs 24.1941μs 41.3325 KOps/s 43.0598 KOps/s $\color{#d91a1a}-4.01\%$
test_step_mdp_speed[False-True-False-False-True] 2.5593ms 23.9483μs 41.7566 KOps/s 43.3486 KOps/s $\color{#d91a1a}-3.67\%$
test_step_mdp_speed[False-True-False-False-False] 48.4010μs 15.4136μs 64.8777 KOps/s 68.0528 KOps/s $\color{#d91a1a}-4.67\%$
test_step_mdp_speed[False-False-True-True-True] 86.2010μs 38.4463μs 26.0103 KOps/s 26.8093 KOps/s $\color{#d91a1a}-2.98\%$
test_step_mdp_speed[False-False-True-True-False] 0.1068ms 26.0537μs 38.3822 KOps/s 40.0491 KOps/s $\color{#d91a1a}-4.16\%$
test_step_mdp_speed[False-False-True-False-True] 67.1450μs 23.4059μs 42.7243 KOps/s 42.4791 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[False-False-True-False-False] 48.8510μs 15.6585μs 63.8630 KOps/s 66.8973 KOps/s $\color{#d91a1a}-4.54\%$
test_step_mdp_speed[False-False-False-True-True] 79.4790μs 39.8031μs 25.1237 KOps/s 25.6684 KOps/s $\color{#d91a1a}-2.12\%$
test_step_mdp_speed[False-False-False-True-False] 73.7280μs 27.6771μs 36.1310 KOps/s 37.6491 KOps/s $\color{#d91a1a}-4.03\%$
test_step_mdp_speed[False-False-False-False-True] 74.7600μs 25.1365μs 39.7827 KOps/s 40.6152 KOps/s $\color{#d91a1a}-2.05\%$
test_step_mdp_speed[False-False-False-False-False] 58.8100μs 17.1746μs 58.2256 KOps/s 60.6270 KOps/s $\color{#d91a1a}-3.96\%$
test_values[generalized_advantage_estimate-True-True] 10.1341ms 9.7848ms 102.1990 Ops/s 102.9360 Ops/s $\color{#d91a1a}-0.72\%$
test_values[vec_generalized_advantage_estimate-True-True] 37.8821ms 35.8095ms 27.9255 Ops/s 27.5280 Ops/s $\color{#35bf28}+1.44\%$
test_values[td0_return_estimate-False-False] 0.2302ms 0.1859ms 5.3783 KOps/s 5.5313 KOps/s $\color{#d91a1a}-2.77\%$
test_values[td1_return_estimate-False-False] 27.9888ms 24.3612ms 41.0488 Ops/s 41.0405 Ops/s $\color{#35bf28}+0.02\%$
test_values[vec_td1_return_estimate-False-False] 38.8899ms 36.0581ms 27.7330 Ops/s 27.4412 Ops/s $\color{#35bf28}+1.06\%$
test_values[td_lambda_return_estimate-True-False] 36.1613ms 34.8640ms 28.6829 Ops/s 28.7058 Ops/s $\color{#d91a1a}-0.08\%$
test_values[vec_td_lambda_return_estimate-True-False] 40.5233ms 35.9050ms 27.8513 Ops/s 27.3790 Ops/s $\color{#35bf28}+1.73\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.9478ms 8.5074ms 117.5448 Ops/s 120.5610 Ops/s $\color{#d91a1a}-2.50\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.2555ms 1.9536ms 511.8858 Ops/s 491.7146 Ops/s $\color{#35bf28}+4.10\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6416ms 0.3597ms 2.7804 KOps/s 2.7472 KOps/s $\color{#35bf28}+1.21\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 48.3987ms 45.3861ms 22.0332 Ops/s 21.0436 Ops/s $\color{#35bf28}+4.70\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.1842ms 3.0601ms 326.7892 Ops/s 325.2825 Ops/s $\color{#35bf28}+0.46\%$
test_dqn_speed[False-None] 5.9181ms 1.3947ms 717.0109 Ops/s 712.6679 Ops/s $\color{#35bf28}+0.61\%$
test_dqn_speed[False-backward] 2.1857ms 1.8827ms 531.1646 Ops/s 521.9490 Ops/s $\color{#35bf28}+1.77\%$
test_dqn_speed[True-None] 0.7080ms 0.4732ms 2.1134 KOps/s 2.1576 KOps/s $\color{#d91a1a}-2.05\%$
test_dqn_speed[True-backward] 1.0289ms 0.8969ms 1.1149 KOps/s 1.1067 KOps/s $\color{#35bf28}+0.74\%$
test_dqn_speed[reduce-overhead-None] 0.6147ms 0.4702ms 2.1266 KOps/s 2.1140 KOps/s $\color{#35bf28}+0.59\%$
test_dqn_speed[reduce-overhead-backward] 1.0501ms 0.9073ms 1.1022 KOps/s 994.1490 Ops/s $\textbf{\color{#35bf28}+10.87\%}$
test_ddpg_speed[False-None] 3.7403ms 2.8895ms 346.0824 Ops/s 345.3350 Ops/s $\color{#35bf28}+0.22\%$
test_ddpg_speed[False-backward] 4.1332ms 4.0267ms 248.3401 Ops/s 250.2784 Ops/s $\color{#d91a1a}-0.77\%$
test_ddpg_speed[True-None] 1.2025ms 1.0058ms 994.2789 Ops/s 982.4204 Ops/s $\color{#35bf28}+1.21\%$
test_ddpg_speed[True-backward] 1.9721ms 1.9067ms 524.4753 Ops/s 509.6146 Ops/s $\color{#35bf28}+2.92\%$
test_ddpg_speed[reduce-overhead-None] 1.1432ms 1.0037ms 996.3260 Ops/s 965.1572 Ops/s $\color{#35bf28}+3.23\%$
test_ddpg_speed[reduce-overhead-backward] 2.1429ms 1.9317ms 517.6760 Ops/s 517.4599 Ops/s $\color{#35bf28}+0.04\%$
test_sac_speed[False-None] 9.6853ms 8.0964ms 123.5121 Ops/s 121.8187 Ops/s $\color{#35bf28}+1.39\%$
test_sac_speed[False-backward] 13.0878ms 10.9755ms 91.1118 Ops/s 87.2979 Ops/s $\color{#35bf28}+4.37\%$
test_sac_speed[True-None] 2.9632ms 1.8448ms 542.0748 Ops/s 511.8296 Ops/s $\textbf{\color{#35bf28}+5.91\%}$
test_sac_speed[True-backward] 3.9645ms 3.5615ms 280.7796 Ops/s 278.3443 Ops/s $\color{#35bf28}+0.87\%$
test_sac_speed[reduce-overhead-None] 2.5426ms 1.8447ms 542.0794 Ops/s 536.7592 Ops/s $\color{#35bf28}+0.99\%$
test_sac_speed[reduce-overhead-backward] 3.6811ms 3.5411ms 282.3991 Ops/s 281.0058 Ops/s $\color{#35bf28}+0.50\%$
test_redq_speed[False-None] 13.5955ms 12.7698ms 78.3098 Ops/s 74.7365 Ops/s $\color{#35bf28}+4.78\%$
test_redq_speed[False-backward] 23.6255ms 22.4463ms 44.5508 Ops/s 43.7751 Ops/s $\color{#35bf28}+1.77\%$
test_redq_speed[True-None] 8.3527ms 5.0584ms 197.6926 Ops/s 199.6004 Ops/s $\color{#d91a1a}-0.96\%$
test_redq_speed[True-backward] 13.1673ms 12.5269ms 79.8282 Ops/s 77.8852 Ops/s $\color{#35bf28}+2.49\%$
test_redq_speed[reduce-overhead-None] 6.4357ms 4.9908ms 200.3705 Ops/s 192.0318 Ops/s $\color{#35bf28}+4.34\%$
test_redq_speed[reduce-overhead-backward] 13.1658ms 12.7650ms 78.3389 Ops/s 77.8176 Ops/s $\color{#35bf28}+0.67\%$
test_redq_deprec_speed[False-None] 14.4691ms 13.4920ms 74.1178 Ops/s 72.0158 Ops/s $\color{#35bf28}+2.92\%$
test_redq_deprec_speed[False-backward] 20.9802ms 19.1727ms 52.1576 Ops/s 34.8201 Ops/s $\textbf{\color{#35bf28}+49.79\%}$
test_redq_deprec_speed[True-None] 4.3159ms 3.7274ms 268.2804 Ops/s 267.0139 Ops/s $\color{#35bf28}+0.47\%$
test_redq_deprec_speed[True-backward] 9.2347ms 8.3383ms 119.9280 Ops/s 118.5534 Ops/s $\color{#35bf28}+1.16\%$
test_redq_deprec_speed[reduce-overhead-None] 5.2272ms 3.6449ms 274.3591 Ops/s 269.5486 Ops/s $\color{#35bf28}+1.78\%$
test_redq_deprec_speed[reduce-overhead-backward] 14.7435ms 8.7196ms 114.6842 Ops/s 118.2956 Ops/s $\color{#d91a1a}-3.05\%$
test_td3_speed[False-None] 33.0372ms 8.2779ms 120.8035 Ops/s 122.0651 Ops/s $\color{#d91a1a}-1.03\%$
test_td3_speed[False-backward] 13.0877ms 10.6127ms 94.2264 Ops/s 92.9129 Ops/s $\color{#35bf28}+1.41\%$
test_td3_speed[True-None] 1.9174ms 1.7106ms 584.5925 Ops/s 573.0626 Ops/s $\color{#35bf28}+2.01\%$
test_td3_speed[True-backward] 3.4736ms 3.3115ms 301.9793 Ops/s 295.3092 Ops/s $\color{#35bf28}+2.26\%$
test_td3_speed[reduce-overhead-None] 1.9649ms 1.6933ms 590.5521 Ops/s 571.7754 Ops/s $\color{#35bf28}+3.28\%$
test_td3_speed[reduce-overhead-backward] 3.6586ms 3.3777ms 296.0575 Ops/s 291.1101 Ops/s $\color{#35bf28}+1.70\%$
test_cql_speed[False-None] 40.0547ms 35.7894ms 27.9413 Ops/s 26.9421 Ops/s $\color{#35bf28}+3.71\%$
test_cql_speed[False-backward] 0.2999s 52.4043ms 19.0824 Ops/s 21.0271 Ops/s $\textbf{\color{#d91a1a}-9.25\%}$
test_cql_speed[True-None] 17.0566ms 16.0036ms 62.4860 Ops/s 62.4730 Ops/s $\color{#35bf28}+0.02\%$
test_cql_speed[True-backward] 23.8838ms 22.9971ms 43.4838 Ops/s 43.8546 Ops/s $\color{#d91a1a}-0.85\%$
test_cql_speed[reduce-overhead-None] 19.3857ms 16.1652ms 61.8613 Ops/s 62.7715 Ops/s $\color{#d91a1a}-1.45\%$
test_cql_speed[reduce-overhead-backward] 23.9018ms 22.9473ms 43.5781 Ops/s 43.7392 Ops/s $\color{#d91a1a}-0.37\%$
test_a2c_speed[False-None] 10.3763ms 7.5709ms 132.0841 Ops/s 133.9166 Ops/s $\color{#d91a1a}-1.37\%$
test_a2c_speed[False-backward] 15.5504ms 15.1054ms 66.2016 Ops/s 66.5507 Ops/s $\color{#d91a1a}-0.52\%$
test_a2c_speed[True-None] 4.7716ms 4.2864ms 233.2963 Ops/s 231.2184 Ops/s $\color{#35bf28}+0.90\%$
test_a2c_speed[True-backward] 11.7251ms 11.1062ms 90.0395 Ops/s 89.7290 Ops/s $\color{#35bf28}+0.35\%$
test_a2c_speed[reduce-overhead-None] 4.9155ms 4.2688ms 234.2593 Ops/s 233.7358 Ops/s $\color{#35bf28}+0.22\%$
test_a2c_speed[reduce-overhead-backward] 11.5628ms 11.1155ms 89.9642 Ops/s 88.4845 Ops/s $\color{#35bf28}+1.67\%$
test_ppo_speed[False-None] 8.5857ms 7.6966ms 129.9278 Ops/s 118.1222 Ops/s $\textbf{\color{#35bf28}+9.99\%}$
test_ppo_speed[False-backward] 17.4186ms 15.4808ms 64.5963 Ops/s 64.7867 Ops/s $\color{#d91a1a}-0.29\%$
test_ppo_speed[True-None] 4.1181ms 3.7412ms 267.2953 Ops/s 253.4086 Ops/s $\textbf{\color{#35bf28}+5.48\%}$
test_ppo_speed[True-backward] 10.5262ms 9.9720ms 100.2804 Ops/s 100.0401 Ops/s $\color{#35bf28}+0.24\%$
test_ppo_speed[reduce-overhead-None] 4.3340ms 3.7406ms 267.3347 Ops/s 265.3498 Ops/s $\color{#35bf28}+0.75\%$
test_ppo_speed[reduce-overhead-backward] 10.6261ms 10.0199ms 99.8017 Ops/s 97.1347 Ops/s $\color{#35bf28}+2.75\%$
test_reinforce_speed[False-None] 7.7047ms 6.6676ms 149.9791 Ops/s 142.8928 Ops/s $\color{#35bf28}+4.96\%$
test_reinforce_speed[False-backward] 10.5260ms 10.2406ms 97.6506 Ops/s 95.0249 Ops/s $\color{#35bf28}+2.76\%$
test_reinforce_speed[True-None] 3.2041ms 2.6850ms 372.4402 Ops/s 356.0318 Ops/s $\color{#35bf28}+4.61\%$
test_reinforce_speed[True-backward] 9.7206ms 8.9987ms 111.1275 Ops/s 111.9019 Ops/s $\color{#d91a1a}-0.69\%$
test_reinforce_speed[reduce-overhead-None] 3.2065ms 2.6916ms 371.5330 Ops/s 368.9050 Ops/s $\color{#35bf28}+0.71\%$
test_reinforce_speed[reduce-overhead-backward] 10.0439ms 8.9933ms 111.1942 Ops/s 112.6932 Ops/s $\color{#d91a1a}-1.33\%$
test_iql_speed[False-None] 34.5867ms 33.0735ms 30.2357 Ops/s 29.6889 Ops/s $\color{#35bf28}+1.84\%$
test_iql_speed[False-backward] 48.7285ms 46.2912ms 21.6024 Ops/s 21.1389 Ops/s $\color{#35bf28}+2.19\%$
test_iql_speed[True-None] 12.4563ms 11.1899ms 89.3666 Ops/s 89.8283 Ops/s $\color{#d91a1a}-0.51\%$
test_iql_speed[True-backward] 23.0554ms 22.3574ms 44.7279 Ops/s 44.2524 Ops/s $\color{#35bf28}+1.07\%$
test_iql_speed[reduce-overhead-None] 12.1802ms 11.0531ms 90.4720 Ops/s 88.8312 Ops/s $\color{#35bf28}+1.85\%$
test_iql_speed[reduce-overhead-backward] 23.3080ms 22.4634ms 44.5169 Ops/s 43.5915 Ops/s $\color{#35bf28}+2.12\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4803ms 5.2476ms 190.5641 Ops/s 184.1765 Ops/s $\color{#35bf28}+3.47\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1259ms 0.5230ms 1.9121 KOps/s 1.8818 KOps/s $\color{#35bf28}+1.61\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7102ms 0.4936ms 2.0260 KOps/s 2.0175 KOps/s $\color{#35bf28}+0.42\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.5481ms 5.0159ms 199.3668 Ops/s 193.8297 Ops/s $\color{#35bf28}+2.86\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8132ms 0.5093ms 1.9636 KOps/s 1.9318 KOps/s $\color{#35bf28}+1.65\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9070ms 0.4823ms 2.0735 KOps/s 2.0590 KOps/s $\color{#35bf28}+0.70\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.0185ms 1.6301ms 613.4663 Ops/s 601.0807 Ops/s $\color{#35bf28}+2.06\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.9256ms 1.5917ms 628.2741 Ops/s 618.0398 Ops/s $\color{#35bf28}+1.66\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.6783ms 5.0645ms 197.4515 Ops/s 193.5670 Ops/s $\color{#35bf28}+2.01\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.3644ms 0.6602ms 1.5148 KOps/s 1.4866 KOps/s $\color{#35bf28}+1.90\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9113ms 0.6360ms 1.5722 KOps/s 1.5100 KOps/s $\color{#35bf28}+4.12\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.5121ms 4.9313ms 202.7870 Ops/s 199.5691 Ops/s $\color{#35bf28}+1.61\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.8824ms 0.5196ms 1.9244 KOps/s 1.8756 KOps/s $\color{#35bf28}+2.60\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7075ms 0.4939ms 2.0247 KOps/s 1.9716 KOps/s $\color{#35bf28}+2.69\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.7805ms 4.8806ms 204.8924 Ops/s 200.5157 Ops/s $\color{#35bf28}+2.18\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7000ms 0.5017ms 1.9933 KOps/s 1.7888 KOps/s $\textbf{\color{#35bf28}+11.44\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 7.3079ms 0.4865ms 2.0554 KOps/s 2.0523 KOps/s $\color{#35bf28}+0.15\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.3524ms 5.0207ms 199.1736 Ops/s 191.2878 Ops/s $\color{#35bf28}+4.12\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1945ms 0.6535ms 1.5303 KOps/s 1.4882 KOps/s $\color{#35bf28}+2.83\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8317ms 0.6306ms 1.5859 KOps/s 1.5606 KOps/s $\color{#35bf28}+1.62\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.0125ms 4.3848ms 228.0608 Ops/s 36.4180 Ops/s $\textbf{\color{#35bf28}+526.23\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.3615ms 2.2096ms 452.5639 Ops/s 392.1156 Ops/s $\textbf{\color{#35bf28}+15.42\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4972ms 1.3152ms 760.3504 Ops/s 767.9035 Ops/s $\color{#d91a1a}-0.98\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4800s 13.9451ms 71.7098 Ops/s 217.0756 Ops/s $\textbf{\color{#d91a1a}-66.97\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.0522ms 2.3316ms 428.8892 Ops/s 423.0760 Ops/s $\color{#35bf28}+1.37\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 4.9236ms 1.2987ms 769.9765 Ops/s 783.0348 Ops/s $\color{#d91a1a}-1.67\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.4534ms 4.6279ms 216.0820 Ops/s 214.9538 Ops/s $\color{#35bf28}+0.52\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.8481ms 2.6297ms 380.2660 Ops/s 365.6207 Ops/s $\color{#35bf28}+4.01\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 4.4525ms 1.4360ms 696.3743 Ops/s 636.7577 Ops/s $\textbf{\color{#35bf28}+9.36\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 11.7189ms 11.0897ms 90.1740 Ops/s 82.4407 Ops/s $\textbf{\color{#35bf28}+9.38\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.2186ms 14.5631ms 68.6669 Ops/s 67.4128 Ops/s $\color{#35bf28}+1.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 20.8014ms 19.8482ms 50.3825 Ops/s 48.5212 Ops/s $\color{#35bf28}+3.84\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.8284ms 14.6864ms 68.0902 Ops/s 67.4498 Ops/s $\color{#35bf28}+0.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.3205ms 20.1104ms 49.7254 Ops/s 49.4806 Ops/s $\color{#35bf28}+0.49\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 17.8196ms 15.8661ms 63.0275 Ops/s 62.5140 Ops/s $\color{#35bf28}+0.82\%$

@github-actions GitHub Actions
Copy link

github-actions bot commented Dec 6, 2024 •
edited
Loading

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7496s 0.7491s 1.3350 Ops/s 1.3053 Ops/s $\color{#35bf28}+2.27\%$
test_transformed 1.1011s 1.0241s 0.9765 Ops/s 1.0027 Ops/s $\color{#d91a1a}-2.62\%$
test_serial 2.2477s 2.1641s 0.4621 Ops/s 0.4688 Ops/s $\color{#d91a1a}-1.43\%$
test_parallel 2.0721s 2.0204s 0.4949 Ops/s 0.5079 Ops/s $\color{#d91a1a}-2.55\%$
test_step_mdp_speed[True-True-True-True-True] 0.1922ms 38.9314μs 25.6862 KOps/s 26.6838 KOps/s $\color{#d91a1a}-3.74\%$
test_step_mdp_speed[True-True-True-True-False] 56.2730μs 22.2874μs 44.8684 KOps/s 45.2716 KOps/s $\color{#d91a1a}-0.89\%$
test_step_mdp_speed[True-True-True-False-True] 63.6030μs 21.6520μs 46.1851 KOps/s 47.7484 KOps/s $\color{#d91a1a}-3.27\%$
test_step_mdp_speed[True-True-True-False-False] 41.5520μs 12.6545μs 79.0232 KOps/s 81.4662 KOps/s $\color{#d91a1a}-3.00\%$
test_step_mdp_speed[True-True-False-True-True] 74.0130μs 40.3362μs 24.7916 KOps/s 24.7983 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[True-True-False-True-False] 49.0520μs 23.7154μs 42.1667 KOps/s 41.8230 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-True-False-False-True] 53.3930μs 24.2017μs 41.3194 KOps/s 44.3208 KOps/s $\textbf{\color{#d91a1a}-6.77\%}$
test_step_mdp_speed[True-True-False-False-False] 48.2620μs 14.6310μs 68.3481 KOps/s 69.6650 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[True-False-True-True-True] 93.9940μs 42.7951μs 23.3671 KOps/s 23.6246 KOps/s $\color{#d91a1a}-1.09\%$
test_step_mdp_speed[True-False-True-True-False] 59.3520μs 25.9835μs 38.4859 KOps/s 38.5163 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[True-False-True-False-True] 72.6030μs 23.1912μs 43.1198 KOps/s 43.8602 KOps/s $\color{#d91a1a}-1.69\%$
test_step_mdp_speed[True-False-True-False-False] 57.7220μs 14.2706μs 70.0742 KOps/s 70.6006 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[True-False-False-True-True] 81.8140μs 44.5814μs 22.4309 KOps/s 22.5108 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-False-False-True-False] 69.0530μs 28.4108μs 35.1979 KOps/s 35.8609 KOps/s $\color{#d91a1a}-1.85\%$
test_step_mdp_speed[True-False-False-False-True] 68.1930μs 25.9374μs 38.5544 KOps/s 40.5236 KOps/s $\color{#d91a1a}-4.86\%$
test_step_mdp_speed[True-False-False-False-False] 53.0030μs 16.5923μs 60.2691 KOps/s 62.1300 KOps/s $\color{#d91a1a}-3.00\%$
test_step_mdp_speed[False-True-True-True-True] 76.1640μs 43.2911μs 23.0994 KOps/s 23.7880 KOps/s $\color{#d91a1a}-2.89\%$
test_step_mdp_speed[False-True-True-True-False] 66.1630μs 26.2330μs 38.1199 KOps/s 39.1048 KOps/s $\color{#d91a1a}-2.52\%$
test_step_mdp_speed[False-True-True-False-True] 55.5630μs 27.4998μs 36.3639 KOps/s 37.1058 KOps/s $\color{#d91a1a}-2.00\%$
test_step_mdp_speed[False-True-True-False-False] 45.0520μs 16.0384μs 62.3503 KOps/s 62.8320 KOps/s $\color{#d91a1a}-0.77\%$
test_step_mdp_speed[False-True-False-True-True] 80.8540μs 44.5053μs 22.4692 KOps/s 23.0538 KOps/s $\color{#d91a1a}-2.54\%$
test_step_mdp_speed[False-True-False-True-False] 66.8940μs 28.3743μs 35.2432 KOps/s 36.1600 KOps/s $\color{#d91a1a}-2.54\%$
test_step_mdp_speed[False-True-False-False-True] 3.1923ms 29.9475μs 33.3918 KOps/s 35.3645 KOps/s $\textbf{\color{#d91a1a}-5.58\%}$
test_step_mdp_speed[False-True-False-False-False] 60.2730μs 18.4705μs 54.1405 KOps/s 56.8646 KOps/s $\color{#d91a1a}-4.79\%$
test_step_mdp_speed[False-False-True-True-True] 79.0140μs 47.8914μs 20.8806 KOps/s 21.9767 KOps/s $\color{#d91a1a}-4.99\%$
test_step_mdp_speed[False-False-True-True-False] 58.0630μs 30.8234μs 32.4428 KOps/s 33.4694 KOps/s $\color{#d91a1a}-3.07\%$
test_step_mdp_speed[False-False-True-False-True] 82.3730μs 29.3164μs 34.1106 KOps/s 35.4850 KOps/s $\color{#d91a1a}-3.87\%$
test_step_mdp_speed[False-False-True-False-False] 79.9940μs 18.0978μs 55.2554 KOps/s 56.0687 KOps/s $\color{#d91a1a}-1.45\%$
test_step_mdp_speed[False-False-False-True-True] 83.5140μs 48.8131μs 20.4863 KOps/s 21.1464 KOps/s $\color{#d91a1a}-3.12\%$
test_step_mdp_speed[False-False-False-True-False] 56.6730μs 32.4238μs 30.8415 KOps/s 31.5352 KOps/s $\color{#d91a1a}-2.20\%$
test_step_mdp_speed[False-False-False-False-True] 62.9830μs 30.4735μs 32.8154 KOps/s 33.5459 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[False-False-False-False-False] 48.1320μs 20.0916μs 49.7720 KOps/s 49.7612 KOps/s $\color{#35bf28}+0.02\%$
test_values[generalized_advantage_estimate-True-True] 25.8396ms 25.4807ms 39.2454 Ops/s 39.5159 Ops/s $\color{#d91a1a}-0.68\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1017s 2.9411ms 340.0036 Ops/s 312.7392 Ops/s $\textbf{\color{#35bf28}+8.72\%}$
test_values[td0_return_estimate-False-False] 0.1063ms 82.5341μs 12.1162 KOps/s 12.0226 KOps/s $\color{#35bf28}+0.78\%$
test_values[td1_return_estimate-False-False] 56.6125ms 56.4140ms 17.7261 Ops/s 17.8725 Ops/s $\color{#d91a1a}-0.82\%$
test_values[vec_td1_return_estimate-False-False] 1.4048ms 1.0976ms 911.1062 Ops/s 913.0494 Ops/s $\color{#d91a1a}-0.21\%$
test_values[td_lambda_return_estimate-True-False] 89.6280ms 89.2705ms 11.2019 Ops/s 11.2496 Ops/s $\color{#d91a1a}-0.42\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4004ms 1.0939ms 914.1248 Ops/s 909.0693 Ops/s $\color{#35bf28}+0.56\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.0722ms 24.9417ms 40.0935 Ops/s 40.0350 Ops/s $\color{#35bf28}+0.15\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0525ms 0.7650ms 1.3073 KOps/s 1.2555 KOps/s $\color{#35bf28}+4.12\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7781ms 0.6830ms 1.4640 KOps/s 1.4649 KOps/s $\color{#d91a1a}-0.06\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5397ms 1.4900ms 671.1610 Ops/s 670.1227 Ops/s $\color{#35bf28}+0.15\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7474ms 0.6972ms 1.4343 KOps/s 1.4343 KOps/s $+0.00\%$
test_dqn_speed[False-None] 7.0172ms 1.5174ms 659.0163 Ops/s 669.2202 Ops/s $\color{#d91a1a}-1.52\%$
test_dqn_speed[False-backward] 2.4965ms 2.1493ms 465.2687 Ops/s 472.3487 Ops/s $\color{#d91a1a}-1.50\%$
test_dqn_speed[True-None] 0.7190ms 0.5273ms 1.8965 KOps/s 1.9122 KOps/s $\color{#d91a1a}-0.82\%$
test_dqn_speed[True-backward] 1.1710ms 1.1116ms 899.6339 Ops/s 839.7961 Ops/s $\textbf{\color{#35bf28}+7.13\%}$
test_dqn_speed[reduce-overhead-None] 0.7285ms 0.5363ms 1.8648 KOps/s 1.8616 KOps/s $\color{#35bf28}+0.17\%$
test_dqn_speed[reduce-overhead-backward] 1.0208ms 0.9542ms 1.0479 KOps/s 936.5615 Ops/s $\textbf{\color{#35bf28}+11.89\%}$
test_ddpg_speed[False-None] 3.2761ms 2.8468ms 351.2731 Ops/s 354.3964 Ops/s $\color{#d91a1a}-0.88\%$
test_ddpg_speed[False-backward] 4.5713ms 4.1050ms 243.6074 Ops/s 238.1205 Ops/s $\color{#35bf28}+2.30\%$
test_ddpg_speed[True-None] 1.1591ms 1.0546ms 948.2574 Ops/s 945.3628 Ops/s $\color{#35bf28}+0.31\%$
test_ddpg_speed[True-backward] 2.2316ms 2.1403ms 467.2193 Ops/s 440.2447 Ops/s $\textbf{\color{#35bf28}+6.13\%}$
test_ddpg_speed[reduce-overhead-None] 1.1853ms 1.0684ms 936.0143 Ops/s 931.0404 Ops/s $\color{#35bf28}+0.53\%$
test_ddpg_speed[reduce-overhead-backward] 1.7449ms 1.6238ms 615.8232 Ops/s 568.2817 Ops/s $\textbf{\color{#35bf28}+8.37\%}$
test_sac_speed[False-None] 8.7367ms 8.1500ms 122.6993 Ops/s 125.3542 Ops/s $\color{#d91a1a}-2.12\%$
test_sac_speed[False-backward] 12.0420ms 11.2185ms 89.1382 Ops/s 89.0069 Ops/s $\color{#35bf28}+0.15\%$
test_sac_speed[True-None] 1.5662ms 1.5006ms 666.3827 Ops/s 661.1247 Ops/s $\color{#35bf28}+0.80\%$
test_sac_speed[True-backward] 3.7181ms 3.1922ms 313.2670 Ops/s 313.6832 Ops/s $\color{#d91a1a}-0.13\%$
test_sac_speed[reduce-overhead-None] 22.2726ms 12.3267ms 81.1245 Ops/s 80.1076 Ops/s $\color{#35bf28}+1.27\%$
test_sac_speed[reduce-overhead-backward] 1.4063ms 1.3299ms 751.9576 Ops/s 667.1401 Ops/s $\textbf{\color{#35bf28}+12.71\%}$
test_redq_speed[False-None] 8.5157ms 7.5883ms 131.7813 Ops/s 132.7316 Ops/s $\color{#d91a1a}-0.72\%$
test_redq_speed[False-backward] 12.1733ms 11.4852ms 87.0688 Ops/s 85.2354 Ops/s $\color{#35bf28}+2.15\%$
test_redq_speed[True-None] 2.0977ms 1.9760ms 506.0704 Ops/s 490.3029 Ops/s $\color{#35bf28}+3.22\%$
test_redq_speed[True-backward] 3.6838ms 3.6016ms 277.6554 Ops/s 274.0558 Ops/s $\color{#35bf28}+1.31\%$
test_redq_speed[reduce-overhead-None] 2.0629ms 1.9712ms 507.3016 Ops/s 503.0214 Ops/s $\color{#35bf28}+0.85\%$
test_redq_speed[reduce-overhead-backward] 4.2267ms 3.7222ms 268.6574 Ops/s 270.3249 Ops/s $\color{#d91a1a}-0.62\%$
test_redq_deprec_speed[False-None] 9.6262ms 9.0780ms 110.1563 Ops/s 110.1547 Ops/s $+0.00\%$
test_redq_deprec_speed[False-backward] 12.9300ms 12.2603ms 81.5644 Ops/s 82.0266 Ops/s $\color{#d91a1a}-0.56\%$
test_redq_deprec_speed[True-None] 2.5265ms 2.4211ms 413.0422 Ops/s 433.3353 Ops/s $\color{#d91a1a}-4.68\%$
test_redq_deprec_speed[True-backward] 4.1751ms 3.9896ms 250.6546 Ops/s 240.7975 Ops/s $\color{#35bf28}+4.09\%$
test_redq_deprec_speed[reduce-overhead-None] 2.3860ms 2.2952ms 435.6992 Ops/s 433.2574 Ops/s $\color{#35bf28}+0.56\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.2043ms 3.9878ms 250.7651 Ops/s 243.0691 Ops/s $\color{#35bf28}+3.17\%$
test_td3_speed[False-None] 7.9722ms 7.8997ms 126.5875 Ops/s 127.3599 Ops/s $\color{#d91a1a}-0.61\%$
test_td3_speed[False-backward] 10.8490ms 10.3486ms 96.6314 Ops/s 95.8617 Ops/s $\color{#35bf28}+0.80\%$
test_td3_speed[True-None] 1.6038ms 1.5426ms 648.2359 Ops/s 649.0211 Ops/s $\color{#d91a1a}-0.12\%$
test_td3_speed[True-backward] 3.0991ms 3.0543ms 327.4065 Ops/s 307.6274 Ops/s $\textbf{\color{#35bf28}+6.43\%}$
test_td3_speed[reduce-overhead-None] 48.8853ms 24.7714ms 40.3692 Ops/s 39.1216 Ops/s $\color{#35bf28}+3.19\%$
test_td3_speed[reduce-overhead-backward] 1.4434ms 1.2960ms 771.5933 Ops/s 693.6922 Ops/s $\textbf{\color{#35bf28}+11.23\%}$
test_cql_speed[False-None] 17.6097ms 16.5662ms 60.3639 Ops/s 60.6784 Ops/s $\color{#d91a1a}-0.52\%$
test_cql_speed[False-backward] 23.0853ms 21.7948ms 45.8825 Ops/s 45.2653 Ops/s $\color{#35bf28}+1.36\%$
test_cql_speed[True-None] 2.9782ms 2.8733ms 348.0278 Ops/s 344.4758 Ops/s $\color{#35bf28}+1.03\%$
test_cql_speed[True-backward] 5.2456ms 4.9890ms 200.4401 Ops/s 191.9430 Ops/s $\color{#35bf28}+4.43\%$
test_cql_speed[reduce-overhead-None] 21.6277ms 13.0493ms 76.6327 Ops/s 77.1269 Ops/s $\color{#d91a1a}-0.64\%$
test_cql_speed[reduce-overhead-backward] 1.5461ms 1.4921ms 670.1887 Ops/s 601.2044 Ops/s $\textbf{\color{#35bf28}+11.47\%}$
test_a2c_speed[False-None] 3.3324ms 3.2249ms 310.0915 Ops/s 308.8031 Ops/s $\color{#35bf28}+0.42\%$
test_a2c_speed[False-backward] 6.6638ms 6.2113ms 160.9979 Ops/s 153.8622 Ops/s $\color{#35bf28}+4.64\%$
test_a2c_speed[True-None] 1.0241ms 0.9819ms 1.0185 KOps/s 1.0043 KOps/s $\color{#35bf28}+1.41\%$
test_a2c_speed[True-backward] 3.1923ms 2.6252ms 380.9282 Ops/s 365.0689 Ops/s $\color{#35bf28}+4.34\%$
test_a2c_speed[reduce-overhead-None] 0.3883s 12.2723ms 81.4844 Ops/s 88.3373 Ops/s $\textbf{\color{#d91a1a}-7.76\%}$
test_a2c_speed[reduce-overhead-backward] 1.0344ms 0.9848ms 1.0155 KOps/s 922.7596 Ops/s $\textbf{\color{#35bf28}+10.05\%}$
test_ppo_speed[False-None] 4.0118ms 3.7377ms 267.5433 Ops/s 268.8061 Ops/s $\color{#d91a1a}-0.47\%$
test_ppo_speed[False-backward] 7.4410ms 7.0120ms 142.6130 Ops/s 140.7961 Ops/s $\color{#35bf28}+1.29\%$
test_ppo_speed[True-None] 1.0240ms 0.9402ms 1.0636 KOps/s 1.0660 KOps/s $\color{#d91a1a}-0.23\%$
test_ppo_speed[True-backward] 2.6601ms 2.5678ms 389.4312 Ops/s 396.0678 Ops/s $\color{#d91a1a}-1.68\%$
test_ppo_speed[reduce-overhead-None] 0.5458ms 0.4828ms 2.0714 KOps/s 1.9249 KOps/s $\textbf{\color{#35bf28}+7.61\%}$
test_ppo_speed[reduce-overhead-backward] 1.1032ms 0.9880ms 1.0122 KOps/s 1.0198 KOps/s $\color{#d91a1a}-0.75\%$
test_reinforce_speed[False-None] 2.5364ms 2.2627ms 441.9584 Ops/s 441.7778 Ops/s $\color{#35bf28}+0.04\%$
test_reinforce_speed[False-backward] 3.4137ms 3.2747ms 305.3725 Ops/s 309.4673 Ops/s $\color{#d91a1a}-1.32\%$
test_reinforce_speed[True-None] 0.8986ms 0.8366ms 1.1953 KOps/s 1.2018 KOps/s $\color{#d91a1a}-0.54\%$
test_reinforce_speed[True-backward] 2.4675ms 2.3896ms 418.4758 Ops/s 413.3224 Ops/s $\color{#35bf28}+1.25\%$
test_reinforce_speed[reduce-overhead-None] 22.4531ms 11.7694ms 84.9660 Ops/s 88.7127 Ops/s $\color{#d91a1a}-4.22\%$
test_reinforce_speed[reduce-overhead-backward] 1.1227ms 1.0573ms 945.8155 Ops/s 949.2598 Ops/s $\color{#d91a1a}-0.36\%$
test_iql_speed[False-None] 9.7159ms 9.2481ms 108.1300 Ops/s 108.9871 Ops/s $\color{#d91a1a}-0.79\%$
test_iql_speed[False-backward] 13.8980ms 13.0808ms 76.4478 Ops/s 77.1705 Ops/s $\color{#d91a1a}-0.94\%$
test_iql_speed[True-None] 1.8608ms 1.7205ms 581.2181 Ops/s 582.6573 Ops/s $\color{#d91a1a}-0.25\%$
test_iql_speed[True-backward] 4.3279ms 4.1686ms 239.8866 Ops/s 224.7020 Ops/s $\textbf{\color{#35bf28}+6.76\%}$
test_iql_speed[reduce-overhead-None] 20.5014ms 11.4029ms 87.6966 Ops/s 89.1262 Ops/s $\color{#d91a1a}-1.60\%$
test_iql_speed[reduce-overhead-backward] 1.4844ms 1.4182ms 705.1144 Ops/s 664.1646 Ops/s $\textbf{\color{#35bf28}+6.17\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7494ms 6.2151ms 160.8993 Ops/s 157.8778 Ops/s $\color{#35bf28}+1.91\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5787ms 0.3277ms 3.0517 KOps/s 3.0350 KOps/s $\color{#35bf28}+0.55\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5314ms 0.3199ms 3.1256 KOps/s 3.1850 KOps/s $\color{#d91a1a}-1.86\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3347ms 6.0299ms 165.8399 Ops/s 164.7265 Ops/s $\color{#35bf28}+0.68\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5131ms 0.2637ms 3.7921 KOps/s 3.3905 KOps/s $\textbf{\color{#35bf28}+11.85\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5821ms 0.2897ms 3.4519 KOps/s 3.9472 KOps/s $\textbf{\color{#d91a1a}-12.55\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4834ms 1.2748ms 784.4224 Ops/s 723.2820 Ops/s $\textbf{\color{#35bf28}+8.45\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5498ms 1.2543ms 797.2691 Ops/s 828.3593 Ops/s $\color{#d91a1a}-3.75\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3272ms 6.1756ms 161.9279 Ops/s 161.0143 Ops/s $\color{#35bf28}+0.57\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1574ms 0.4435ms 2.2547 KOps/s 2.1160 KOps/s $\textbf{\color{#35bf28}+6.56\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7122ms 0.4227ms 2.3657 KOps/s 2.2909 KOps/s $\color{#35bf28}+3.26\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1368ms 6.0354ms 165.6891 Ops/s 164.5248 Ops/s $\color{#35bf28}+0.71\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0294ms 0.3501ms 2.8562 KOps/s 3.6045 KOps/s $\textbf{\color{#d91a1a}-20.76\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5370ms 0.2977ms 3.3588 KOps/s 3.0841 KOps/s $\textbf{\color{#35bf28}+8.91\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2023ms 5.9705ms 167.4913 Ops/s 167.4580 Ops/s $\color{#35bf28}+0.02\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.5916ms 0.2827ms 3.5378 KOps/s 2.7930 KOps/s $\textbf{\color{#35bf28}+26.67\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5134ms 0.2616ms 3.8224 KOps/s 4.0779 KOps/s $\textbf{\color{#d91a1a}-6.27\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3575ms 6.2271ms 160.5885 Ops/s 161.2404 Ops/s $\color{#d91a1a}-0.40\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.3752ms 0.5005ms 1.9978 KOps/s 2.1818 KOps/s $\textbf{\color{#d91a1a}-8.43\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6776ms 0.4882ms 2.0483 KOps/s 2.1807 KOps/s $\textbf{\color{#d91a1a}-6.07\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0927ms 5.4442ms 183.6821 Ops/s 187.4479 Ops/s $\color{#d91a1a}-2.01\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.2915ms 2.0760ms 481.7050 Ops/s 512.3786 Ops/s $\textbf{\color{#d91a1a}-5.99\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.6039ms 1.1904ms 840.0622 Ops/s 873.2310 Ops/s $\color{#d91a1a}-3.80\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5001s 15.3738ms 65.0456 Ops/s 192.0204 Ops/s $\textbf{\color{#d91a1a}-66.13\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 6.4748ms 2.0082ms 497.9685 Ops/s 432.1075 Ops/s $\textbf{\color{#35bf28}+15.24\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 9.8906ms 1.2687ms 788.2348 Ops/s 856.3558 Ops/s $\textbf{\color{#d91a1a}-7.95\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 9.6165ms 5.7689ms 173.3436 Ops/s 33.3734 Ops/s $\textbf{\color{#35bf28}+419.41\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.5009ms 2.2079ms 452.9162 Ops/s 510.3938 Ops/s $\textbf{\color{#d91a1a}-11.26\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.5246ms 1.4064ms 711.0104 Ops/s 821.7173 Ops/s $\textbf{\color{#d91a1a}-13.47\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 17.6672ms 12.9630ms 77.1426 Ops/s 77.8086 Ops/s $\color{#d91a1a}-0.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 17.3558ms 16.8308ms 59.4149 Ops/s 60.8714 Ops/s $\color{#d91a1a}-2.39\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.0134ms 17.5299ms 57.0454 Ops/s 55.7109 Ops/s $\color{#35bf28}+2.40\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 17.5284ms 16.9332ms 59.0555 Ops/s 59.9004 Ops/s $\color{#d91a1a}-1.41\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.4771ms 17.8329ms 56.0760 Ops/s 56.5616 Ops/s $\color{#d91a1a}-0.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 18.7079ms 18.1130ms 55.2090 Ops/s 56.1212 Ops/s $\color{#d91a1a}-1.63\%$

@vmoens
Update
ae2bc40 
[ghstack-poisoned]
@vmoens
Update
a874ea2 
[ghstack-poisoned]
@vmoens
Update
b827784 
[ghstack-poisoned]
@vmoens vmoens merged commit b827784 into gh/vmoens/52/base Dec 12, 2024
1 check passed
vmoens added a commit that referenced this pull request Dec 12, 2024
@vmoens
[Feature] spec.cardinality
dd26ae7 
ghstack-source-id: 1160900f8a81dd51dc72436e1af69c8248bff162
Pull Request resolved: #2638
@vmoens vmoens deleted the gh/vmoens/52/head branch December 12, 2024 21:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
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
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vmoens @facebook-github-bot