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,NOMRERG] Test @v4 #2686

Closed
wants to merge 3 commits into from
Closed

[CI,NOMRERG] Test @v4 #2686

wants to merge 3 commits into from

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Jan 9, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 9, 2025
ghstack-source-id: 23cd637b1c49239ce9cd673a03bccebf03a4a612
Pull Request resolved: #2686
Copy link

pytorch-bot bot commented Jan 9, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2686

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

❌ 12 New Failures

As of commit 051d999 with merge base f672c70 (image):

NEW FAILURES - The following jobs have failed:

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 Jan 9, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 9, 2025
ghstack-source-id: 208476aa44cca5c1b8feaac193ce22c4cce3da4d
Pull Request resolved: #2686
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Jan 9, 2025
ghstack-source-id: b865bfa51797ffc08dd75734edcd5f082a8b6def
Pull Request resolved: #2686
Copy link

github-actions bot commented Jan 9, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}17$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5490s 0.4535s 2.2051 Ops/s 2.1773 Ops/s $\color{#35bf28}+1.28\%$
test_transformed 0.7334s 0.6368s 1.5704 Ops/s 1.5706 Ops/s $\color{#d91a1a}-0.01\%$
test_serial 1.4954s 1.4044s 0.7121 Ops/s 0.7221 Ops/s $\color{#d91a1a}-1.39\%$
test_parallel 1.3193s 1.2221s 0.8183 Ops/s 0.8152 Ops/s $\color{#35bf28}+0.38\%$
test_step_mdp_speed[True-True-True-True-True] 0.2818ms 30.2755μs 33.0300 KOps/s 32.4943 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-True-True-True-False] 56.0040μs 17.7757μs 56.2567 KOps/s 55.3786 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-True-True-False-True] 71.7460μs 16.9102μs 59.1358 KOps/s 58.4648 KOps/s $\color{#35bf28}+1.15\%$
test_step_mdp_speed[True-True-True-False-False] 48.8220μs 9.8977μs 101.0331 KOps/s 98.7910 KOps/s $\color{#35bf28}+2.27\%$
test_step_mdp_speed[True-True-False-True-True] 75.3510μs 31.9882μs 31.2615 KOps/s 30.7006 KOps/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[True-True-False-True-False] 70.7530μs 19.6334μs 50.9335 KOps/s 50.0151 KOps/s $\color{#35bf28}+1.84\%$
test_step_mdp_speed[True-True-False-False-True] 59.7420μs 19.0607μs 52.4640 KOps/s 52.2442 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[True-True-False-False-False] 58.7400μs 11.8110μs 84.6667 KOps/s 82.8891 KOps/s $\color{#35bf28}+2.14\%$
test_step_mdp_speed[True-False-True-True-True] 87.1330μs 33.8627μs 29.5310 KOps/s 28.7091 KOps/s $\color{#35bf28}+2.86\%$
test_step_mdp_speed[True-False-True-True-False] 72.1450μs 21.4371μs 46.6482 KOps/s 45.8250 KOps/s $\color{#35bf28}+1.80\%$
test_step_mdp_speed[True-False-True-False-True] 60.8740μs 18.8183μs 53.1398 KOps/s 52.4657 KOps/s $\color{#35bf28}+1.28\%$
test_step_mdp_speed[True-False-True-False-False] 49.4820μs 11.7675μs 84.9800 KOps/s 83.1668 KOps/s $\color{#35bf28}+2.18\%$
test_step_mdp_speed[True-False-False-True-True] 86.3620μs 35.8254μs 27.9132 KOps/s 27.8204 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[True-False-False-True-False] 66.5250μs 23.1079μs 43.2753 KOps/s 42.1159 KOps/s $\color{#35bf28}+2.75\%$
test_step_mdp_speed[True-False-False-False-True] 0.5887ms 20.6104μs 48.5192 KOps/s 48.2150 KOps/s $\color{#35bf28}+0.63\%$
test_step_mdp_speed[True-False-False-False-False] 53.7010μs 13.5703μs 73.6903 KOps/s 71.7539 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[False-True-True-True-True] 84.5990μs 34.1378μs 29.2930 KOps/s 29.0566 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[False-True-True-True-False] 58.5400μs 21.4501μs 46.6199 KOps/s 45.7299 KOps/s $\color{#35bf28}+1.95\%$
test_step_mdp_speed[False-True-True-False-True] 67.6970μs 21.8476μs 45.7717 KOps/s 45.3458 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[False-True-True-False-False] 51.3070μs 13.3935μs 74.6630 KOps/s 72.9147 KOps/s $\color{#35bf28}+2.40\%$
test_step_mdp_speed[False-True-False-True-True] 76.4230μs 35.8289μs 27.9105 KOps/s 27.6827 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-True-False-True-False] 63.6800μs 23.1654μs 43.1678 KOps/s 41.9187 KOps/s $\color{#35bf28}+2.98\%$
test_step_mdp_speed[False-True-False-False-True] 2.6582ms 23.6033μs 42.3669 KOps/s 41.5813 KOps/s $\color{#35bf28}+1.89\%$
test_step_mdp_speed[False-True-False-False-False] 59.4210μs 15.0654μs 66.3771 KOps/s 60.4978 KOps/s $\textbf{\color{#35bf28}+9.72\%}$
test_step_mdp_speed[False-False-True-True-True] 0.1215ms 37.6500μs 26.5605 KOps/s 26.3789 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[False-False-True-True-False] 72.0150μs 25.1472μs 39.7658 KOps/s 39.0692 KOps/s $\color{#35bf28}+1.78\%$
test_step_mdp_speed[False-False-True-False-True] 69.0000μs 23.4779μs 42.5932 KOps/s 42.0192 KOps/s $\color{#35bf28}+1.37\%$
test_step_mdp_speed[False-False-True-False-False] 53.3100μs 15.2203μs 65.7017 KOps/s 65.2796 KOps/s $\color{#35bf28}+0.65\%$
test_step_mdp_speed[False-False-False-True-True] 79.5800μs 39.5696μs 25.2719 KOps/s 25.2199 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[False-False-False-True-False] 72.0750μs 27.1104μs 36.8862 KOps/s 36.6092 KOps/s $\color{#35bf28}+0.76\%$
test_step_mdp_speed[False-False-False-False-True] 63.2990μs 24.8802μs 40.1927 KOps/s 39.3370 KOps/s $\color{#35bf28}+2.18\%$
test_step_mdp_speed[False-False-False-False-False] 66.2140μs 16.8883μs 59.2126 KOps/s 58.5160 KOps/s $\color{#35bf28}+1.19\%$
test_values[generalized_advantage_estimate-True-True] 11.0899ms 9.9832ms 100.1684 Ops/s 98.0913 Ops/s $\color{#35bf28}+2.12\%$
test_values[vec_generalized_advantage_estimate-True-True] 43.6524ms 34.9646ms 28.6003 Ops/s 29.4267 Ops/s $\color{#d91a1a}-2.81\%$
test_values[td0_return_estimate-False-False] 0.2364ms 0.2026ms 4.9354 KOps/s 4.5498 KOps/s $\textbf{\color{#35bf28}+8.47\%}$
test_values[td1_return_estimate-False-False] 27.8839ms 24.6376ms 40.5884 Ops/s 39.6248 Ops/s $\color{#35bf28}+2.43\%$
test_values[vec_td1_return_estimate-False-False] 49.7526ms 35.9517ms 27.8151 Ops/s 28.3404 Ops/s $\color{#d91a1a}-1.85\%$
test_values[td_lambda_return_estimate-True-False] 38.1128ms 35.0602ms 28.5223 Ops/s 27.5366 Ops/s $\color{#35bf28}+3.58\%$
test_values[vec_td_lambda_return_estimate-True-False] 37.3336ms 35.0710ms 28.5136 Ops/s 28.0834 Ops/s $\color{#35bf28}+1.53\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 15.7087ms 8.6908ms 115.0640 Ops/s 115.9546 Ops/s $\color{#d91a1a}-0.77\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.5537ms 1.8901ms 529.0799 Ops/s 554.8177 Ops/s $\color{#d91a1a}-4.64\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6048ms 0.3613ms 2.7680 KOps/s 2.7542 KOps/s $\color{#35bf28}+0.50\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 43.3912ms 41.9568ms 23.8341 Ops/s 23.0648 Ops/s $\color{#35bf28}+3.34\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.3545ms 3.0977ms 322.8175 Ops/s 327.4707 Ops/s $\color{#d91a1a}-1.42\%$
test_dqn_speed[False-None] 1.9074ms 1.4150ms 706.7171 Ops/s 698.4293 Ops/s $\color{#35bf28}+1.19\%$
test_dqn_speed[False-backward] 1.9688ms 1.9012ms 525.9923 Ops/s 501.5255 Ops/s $\color{#35bf28}+4.88\%$
test_dqn_speed[True-None] 0.8695ms 0.4801ms 2.0830 KOps/s 2.0238 KOps/s $\color{#35bf28}+2.92\%$
test_dqn_speed[True-backward] 1.0279ms 0.9171ms 1.0904 KOps/s 987.1184 Ops/s $\textbf{\color{#35bf28}+10.46\%}$
test_dqn_speed[reduce-overhead-None] 0.8283ms 0.4853ms 2.0604 KOps/s 2.0365 KOps/s $\color{#35bf28}+1.18\%$
test_dqn_speed[reduce-overhead-backward] 0.9823ms 0.9099ms 1.0990 KOps/s 1.0694 KOps/s $\color{#35bf28}+2.77\%$
test_ddpg_speed[False-None] 6.4566ms 3.1326ms 319.2274 Ops/s 342.1874 Ops/s $\textbf{\color{#d91a1a}-6.71\%}$
test_ddpg_speed[False-backward] 4.6420ms 4.2626ms 234.5985 Ops/s 242.8553 Ops/s $\color{#d91a1a}-3.40\%$
test_ddpg_speed[True-None] 2.0737ms 1.0326ms 968.4658 Ops/s 964.5746 Ops/s $\color{#35bf28}+0.40\%$
test_ddpg_speed[True-backward] 2.1755ms 1.9898ms 502.5536 Ops/s 514.1178 Ops/s $\color{#d91a1a}-2.25\%$
test_ddpg_speed[reduce-overhead-None] 1.2293ms 1.0265ms 974.1745 Ops/s 950.8589 Ops/s $\color{#35bf28}+2.45\%$
test_ddpg_speed[reduce-overhead-backward] 2.1406ms 1.9999ms 500.0290 Ops/s 451.7015 Ops/s $\textbf{\color{#35bf28}+10.70\%}$
test_sac_speed[False-None] 9.2877ms 8.5955ms 116.3403 Ops/s 122.1025 Ops/s $\color{#d91a1a}-4.72\%$
test_sac_speed[False-backward] 14.0937ms 11.9673ms 83.5608 Ops/s 90.4358 Ops/s $\textbf{\color{#d91a1a}-7.60\%}$
test_sac_speed[True-None] 2.6876ms 1.9156ms 522.0395 Ops/s 524.0972 Ops/s $\color{#d91a1a}-0.39\%$
test_sac_speed[True-backward] 4.2273ms 4.0329ms 247.9627 Ops/s 247.9517 Ops/s $+0.00\%$
test_sac_speed[reduce-overhead-None] 2.5199ms 1.9532ms 511.9776 Ops/s 523.0599 Ops/s $\color{#d91a1a}-2.12\%$
test_sac_speed[reduce-overhead-backward] 4.2746ms 3.9452ms 253.4697 Ops/s 237.4127 Ops/s $\textbf{\color{#35bf28}+6.76\%}$
test_redq_speed[False-None] 15.3766ms 13.5998ms 73.5308 Ops/s 70.8903 Ops/s $\color{#35bf28}+3.72\%$
test_redq_speed[False-backward] 24.8117ms 23.1154ms 43.2612 Ops/s 41.7640 Ops/s $\color{#35bf28}+3.59\%$
test_redq_speed[True-None] 6.4586ms 5.6529ms 176.8994 Ops/s 189.4360 Ops/s $\textbf{\color{#d91a1a}-6.62\%}$
test_redq_speed[True-backward] 14.8245ms 13.1138ms 76.2557 Ops/s 75.9543 Ops/s $\color{#35bf28}+0.40\%$
test_redq_speed[reduce-overhead-None] 6.6402ms 5.7283ms 174.5729 Ops/s 181.3542 Ops/s $\color{#d91a1a}-3.74\%$
test_redq_speed[reduce-overhead-backward] 14.0875ms 12.8590ms 77.7665 Ops/s 77.2110 Ops/s $\color{#35bf28}+0.72\%$
test_redq_deprec_speed[False-None] 16.0845ms 14.3207ms 69.8288 Ops/s 71.0369 Ops/s $\color{#d91a1a}-1.70\%$
test_redq_deprec_speed[False-backward] 21.7510ms 19.9677ms 50.0808 Ops/s 49.3658 Ops/s $\color{#35bf28}+1.45\%$
test_redq_deprec_speed[True-None] 4.8205ms 4.3988ms 227.3357 Ops/s 245.1433 Ops/s $\textbf{\color{#d91a1a}-7.26\%}$
test_redq_deprec_speed[True-backward] 10.1822ms 9.2087ms 108.5924 Ops/s 109.2792 Ops/s $\color{#d91a1a}-0.63\%$
test_redq_deprec_speed[reduce-overhead-None] 4.8244ms 4.3113ms 231.9491 Ops/s 224.1221 Ops/s $\color{#35bf28}+3.49\%$
test_redq_deprec_speed[reduce-overhead-backward] 10.3906ms 9.2091ms 108.5881 Ops/s 108.5924 Ops/s $-0.00\%$
test_td3_speed[False-None] 9.2694ms 8.7003ms 114.9381 Ops/s 112.6463 Ops/s $\color{#35bf28}+2.03\%$
test_td3_speed[False-backward] 13.5260ms 11.5168ms 86.8297 Ops/s 89.1247 Ops/s $\color{#d91a1a}-2.57\%$
test_td3_speed[True-None] 1.9619ms 1.8019ms 554.9615 Ops/s 544.5507 Ops/s $\color{#35bf28}+1.91\%$
test_td3_speed[True-backward] 4.2014ms 3.7932ms 263.6308 Ops/s 275.0503 Ops/s $\color{#d91a1a}-4.15\%$
test_td3_speed[reduce-overhead-None] 3.1750ms 1.8896ms 529.2166 Ops/s 535.2247 Ops/s $\color{#d91a1a}-1.12\%$
test_td3_speed[reduce-overhead-backward] 4.1261ms 3.8644ms 258.7693 Ops/s 259.1396 Ops/s $\color{#d91a1a}-0.14\%$
test_cql_speed[False-None] 41.6411ms 37.9845ms 26.3265 Ops/s 25.7802 Ops/s $\color{#35bf28}+2.12\%$
test_cql_speed[False-backward] 52.2464ms 49.4996ms 20.2022 Ops/s 20.2816 Ops/s $\color{#d91a1a}-0.39\%$
test_cql_speed[True-None] 17.6191ms 16.3127ms 61.3021 Ops/s 59.2941 Ops/s $\color{#35bf28}+3.39\%$
test_cql_speed[True-backward] 43.6233ms 24.5758ms 40.6905 Ops/s 40.7095 Ops/s $\color{#d91a1a}-0.05\%$
test_cql_speed[reduce-overhead-None] 17.4457ms 16.5488ms 60.4275 Ops/s 60.0971 Ops/s $\color{#35bf28}+0.55\%$
test_cql_speed[reduce-overhead-backward] 24.3967ms 23.5139ms 42.5280 Ops/s 41.8724 Ops/s $\color{#35bf28}+1.57\%$
test_a2c_speed[False-None] 9.2851ms 8.0267ms 124.5848 Ops/s 120.4526 Ops/s $\color{#35bf28}+3.43\%$
test_a2c_speed[False-backward] 15.7563ms 15.4322ms 64.7997 Ops/s 62.8876 Ops/s $\color{#35bf28}+3.04\%$
test_a2c_speed[True-None] 5.2942ms 4.6878ms 213.3196 Ops/s 217.9106 Ops/s $\color{#d91a1a}-2.11\%$
test_a2c_speed[True-backward] 12.1603ms 11.8289ms 84.5385 Ops/s 87.5145 Ops/s $\color{#d91a1a}-3.40\%$
test_a2c_speed[reduce-overhead-None] 5.5730ms 4.7758ms 209.3881 Ops/s 225.5802 Ops/s $\textbf{\color{#d91a1a}-7.18\%}$
test_a2c_speed[reduce-overhead-backward] 12.8409ms 11.6354ms 85.9444 Ops/s 87.3917 Ops/s $\color{#d91a1a}-1.66\%$
test_ppo_speed[False-None] 9.1899ms 8.3442ms 119.8430 Ops/s 119.8235 Ops/s $\color{#35bf28}+0.02\%$
test_ppo_speed[False-backward] 17.1272ms 15.8326ms 63.1609 Ops/s 62.9044 Ops/s $\color{#35bf28}+0.41\%$
test_ppo_speed[True-None] 4.3902ms 4.0178ms 248.8953 Ops/s 244.9309 Ops/s $\color{#35bf28}+1.62\%$
test_ppo_speed[True-backward] 10.6895ms 10.4604ms 95.5990 Ops/s 95.0579 Ops/s $\color{#35bf28}+0.57\%$
test_ppo_speed[reduce-overhead-None] 9.7383ms 4.4017ms 227.1857 Ops/s 252.1707 Ops/s $\textbf{\color{#d91a1a}-9.91\%}$
test_ppo_speed[reduce-overhead-backward] 10.6635ms 10.3702ms 96.4303 Ops/s 97.2644 Ops/s $\color{#d91a1a}-0.86\%$
test_reinforce_speed[False-None] 8.2652ms 6.9085ms 144.7499 Ops/s 144.2518 Ops/s $\color{#35bf28}+0.35\%$
test_reinforce_speed[False-backward] 20.2456ms 10.7172ms 93.3083 Ops/s 97.5407 Ops/s $\color{#d91a1a}-4.34\%$
test_reinforce_speed[True-None] 4.3577ms 3.0012ms 333.1968 Ops/s 356.1008 Ops/s $\textbf{\color{#d91a1a}-6.43\%}$
test_reinforce_speed[True-backward] 11.6805ms 9.5577ms 104.6279 Ops/s 109.5984 Ops/s $\color{#d91a1a}-4.54\%$
test_reinforce_speed[reduce-overhead-None] 4.6076ms 3.1244ms 320.0598 Ops/s 360.0070 Ops/s $\textbf{\color{#d91a1a}-11.10\%}$
test_reinforce_speed[reduce-overhead-backward] 9.8255ms 9.4046ms 106.3309 Ops/s 110.8334 Ops/s $\color{#d91a1a}-4.06\%$
test_iql_speed[False-None] 40.9623ms 34.5874ms 28.9122 Ops/s 29.8156 Ops/s $\color{#d91a1a}-3.03\%$
test_iql_speed[False-backward] 57.6924ms 48.2375ms 20.7308 Ops/s 21.2997 Ops/s $\color{#d91a1a}-2.67\%$
test_iql_speed[True-None] 12.3561ms 11.3984ms 87.7319 Ops/s 87.6224 Ops/s $\color{#35bf28}+0.12\%$
test_iql_speed[True-backward] 24.3367ms 22.9437ms 43.5850 Ops/s 43.7596 Ops/s $\color{#d91a1a}-0.40\%$
test_iql_speed[reduce-overhead-None] 12.0485ms 11.3323ms 88.2431 Ops/s 88.1465 Ops/s $\color{#35bf28}+0.11\%$
test_iql_speed[reduce-overhead-backward] 24.0254ms 23.0341ms 43.4139 Ops/s 44.7014 Ops/s $\color{#d91a1a}-2.88\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0134ms 5.5968ms 178.6720 Ops/s 188.3143 Ops/s $\textbf{\color{#d91a1a}-5.12\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8652ms 0.5469ms 1.8284 KOps/s 666.5816 Ops/s $\textbf{\color{#35bf28}+174.29\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9512ms 0.5237ms 1.9095 KOps/s 1.9583 KOps/s $\color{#d91a1a}-2.49\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.6177ms 5.2998ms 188.6881 Ops/s 200.4495 Ops/s $\textbf{\color{#d91a1a}-5.87\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6754ms 0.5354ms 1.8679 KOps/s 1.9512 KOps/s $\color{#d91a1a}-4.27\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8198ms 0.5119ms 1.9534 KOps/s 1.9343 KOps/s $\color{#35bf28}+0.99\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.5021ms 1.6804ms 595.1021 Ops/s 579.8592 Ops/s $\color{#35bf28}+2.63\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.3404ms 1.6005ms 624.7863 Ops/s 628.4617 Ops/s $\color{#d91a1a}-0.58\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.4718ms 5.5303ms 180.8214 Ops/s 194.3884 Ops/s $\textbf{\color{#d91a1a}-6.98\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 4.3771ms 0.6797ms 1.4713 KOps/s 1.4974 KOps/s $\color{#d91a1a}-1.74\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8971ms 0.6474ms 1.5446 KOps/s 1.5717 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8200ms 5.2130ms 191.8296 Ops/s 196.3908 Ops/s $\color{#d91a1a}-2.32\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.4861ms 0.5494ms 1.8201 KOps/s 1.8398 KOps/s $\color{#d91a1a}-1.07\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7621ms 0.5203ms 1.9220 KOps/s 1.9616 KOps/s $\color{#d91a1a}-2.02\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9608ms 5.1812ms 193.0039 Ops/s 202.4352 Ops/s $\color{#d91a1a}-4.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.3022ms 0.5322ms 1.8789 KOps/s 1.9444 KOps/s $\color{#d91a1a}-3.37\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8651ms 0.5119ms 1.9536 KOps/s 1.9992 KOps/s $\color{#d91a1a}-2.28\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8648ms 5.4491ms 183.5177 Ops/s 195.0614 Ops/s $\textbf{\color{#d91a1a}-5.92\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2311ms 0.6772ms 1.4767 KOps/s 1.5251 KOps/s $\color{#d91a1a}-3.17\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9621ms 0.6550ms 1.5267 KOps/s 1.5686 KOps/s $\color{#d91a1a}-2.67\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5185s 14.9342ms 66.9604 Ops/s 245.3974 Ops/s $\textbf{\color{#d91a1a}-72.71\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.9279ms 2.4126ms 414.4865 Ops/s 427.3130 Ops/s $\color{#d91a1a}-3.00\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.7715ms 1.3449ms 743.5678 Ops/s 733.1082 Ops/s $\color{#35bf28}+1.43\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.0446ms 4.7212ms 211.8115 Ops/s 251.6087 Ops/s $\textbf{\color{#d91a1a}-15.82\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.7689ms 2.3780ms 420.5262 Ops/s 447.5188 Ops/s $\textbf{\color{#d91a1a}-6.03\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.7542ms 1.3417ms 745.3088 Ops/s 628.9406 Ops/s $\textbf{\color{#35bf28}+18.50\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4565s 13.8688ms 72.1044 Ops/s 241.5006 Ops/s $\textbf{\color{#d91a1a}-70.14\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 7.5777ms 2.5116ms 398.1474 Ops/s 420.9915 Ops/s $\textbf{\color{#d91a1a}-5.43\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.0203ms 1.5145ms 660.2991 Ops/s 684.8632 Ops/s $\color{#d91a1a}-3.59\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 14.8732ms 13.5042ms 74.0511 Ops/s 74.2735 Ops/s $\color{#d91a1a}-0.30\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.6887ms 15.2167ms 65.7172 Ops/s 66.9717 Ops/s $\color{#d91a1a}-1.87\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 23.4755ms 22.3621ms 44.7185 Ops/s 45.1925 Ops/s $\color{#d91a1a}-1.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 17.1602ms 15.4400ms 64.7667 Ops/s 63.3741 Ops/s $\color{#35bf28}+2.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 22.7216ms 22.1621ms 45.1220 Ops/s 44.8151 Ops/s $\color{#35bf28}+0.68\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 18.2170ms 16.7480ms 59.7085 Ops/s 60.8228 Ops/s $\color{#d91a1a}-1.83\%$

Copy link

github-actions bot commented Jan 9, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.6992s 0.6985s 1.4316 Ops/s 1.3908 Ops/s $\color{#35bf28}+2.94\%$
test_transformed 0.9464s 0.9447s 1.0585 Ops/s 1.0373 Ops/s $\color{#35bf28}+2.04\%$
test_serial 2.1667s 2.0815s 0.4804 Ops/s 0.4814 Ops/s $\color{#d91a1a}-0.20\%$
test_parallel 1.9280s 1.8475s 0.5413 Ops/s 0.5447 Ops/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[True-True-True-True-True] 0.2310ms 38.1422μs 26.2177 KOps/s 26.8026 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[True-True-True-True-False] 54.3410μs 22.1274μs 45.1928 KOps/s 45.6453 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[True-True-True-False-True] 51.6110μs 21.3795μs 46.7737 KOps/s 47.6159 KOps/s $\color{#d91a1a}-1.77\%$
test_step_mdp_speed[True-True-True-False-False] 53.0110μs 12.3274μs 81.1201 KOps/s 80.9040 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-True-False-True-True] 69.5410μs 40.8214μs 24.4969 KOps/s 24.3967 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-True-False-True-False] 55.1210μs 24.5964μs 40.6564 KOps/s 40.9934 KOps/s $\color{#d91a1a}-0.82\%$
test_step_mdp_speed[True-True-False-False-True] 51.2710μs 23.3928μs 42.7481 KOps/s 42.3103 KOps/s $\color{#35bf28}+1.03\%$
test_step_mdp_speed[True-True-False-False-False] 44.7310μs 14.7346μs 67.8675 KOps/s 69.3794 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[True-False-True-True-True] 74.1910μs 43.0579μs 23.2245 KOps/s 23.3368 KOps/s $\color{#d91a1a}-0.48\%$
test_step_mdp_speed[True-False-True-True-False] 59.7310μs 26.5889μs 37.6096 KOps/s 36.9896 KOps/s $\color{#35bf28}+1.68\%$
test_step_mdp_speed[True-False-True-False-True] 54.3910μs 23.7235μs 42.1524 KOps/s 43.8539 KOps/s $\color{#d91a1a}-3.88\%$
test_step_mdp_speed[True-False-True-False-False] 45.1700μs 14.5763μs 68.6044 KOps/s 67.9901 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[True-False-False-True-True] 80.8310μs 45.0708μs 22.1873 KOps/s 22.3755 KOps/s $\color{#d91a1a}-0.84\%$
test_step_mdp_speed[True-False-False-True-False] 64.4700μs 28.5586μs 35.0157 KOps/s 34.7332 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[True-False-False-False-True] 54.0910μs 25.4508μs 39.2915 KOps/s 39.7999 KOps/s $\color{#d91a1a}-1.28\%$
test_step_mdp_speed[True-False-False-False-False] 49.3700μs 16.4031μs 60.9641 KOps/s 59.2108 KOps/s $\color{#35bf28}+2.96\%$
test_step_mdp_speed[False-True-True-True-True] 84.3520μs 42.5962μs 23.4763 KOps/s 24.1126 KOps/s $\color{#d91a1a}-2.64\%$
test_step_mdp_speed[False-True-True-True-False] 57.9110μs 26.1216μs 38.2825 KOps/s 38.6317 KOps/s $\color{#d91a1a}-0.90\%$
test_step_mdp_speed[False-True-True-False-True] 60.1910μs 26.9372μs 37.1234 KOps/s 37.7024 KOps/s $\color{#d91a1a}-1.54\%$
test_step_mdp_speed[False-True-True-False-False] 44.4210μs 16.1885μs 61.7724 KOps/s 64.0310 KOps/s $\color{#d91a1a}-3.53\%$
test_step_mdp_speed[False-True-False-True-True] 80.2610μs 45.3734μs 22.0394 KOps/s 22.9982 KOps/s $\color{#d91a1a}-4.17\%$
test_step_mdp_speed[False-True-False-True-False] 57.8110μs 29.2052μs 34.2405 KOps/s 34.9079 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[False-True-False-False-True] 3.2659ms 29.9766μs 33.3594 KOps/s 33.5513 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[False-True-False-False-False] 51.6210μs 18.4853μs 54.0972 KOps/s 53.6131 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[False-False-True-True-True] 74.0010μs 47.0855μs 21.2380 KOps/s 21.2055 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[False-False-True-True-False] 67.9410μs 30.8268μs 32.4393 KOps/s 32.0354 KOps/s $\color{#35bf28}+1.26\%$
test_step_mdp_speed[False-False-True-False-True] 60.6710μs 29.6339μs 33.7452 KOps/s 34.7432 KOps/s $\color{#d91a1a}-2.87\%$
test_step_mdp_speed[False-False-True-False-False] 47.8710μs 18.7563μs 53.3155 KOps/s 56.9297 KOps/s $\textbf{\color{#d91a1a}-6.35\%}$
test_step_mdp_speed[False-False-False-True-True] 80.3510μs 49.6748μs 20.1309 KOps/s 20.4303 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-False-False-True-False] 66.6910μs 33.1590μs 30.1577 KOps/s 30.0482 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[False-False-False-False-True] 59.2310μs 30.9026μs 32.3598 KOps/s 32.3796 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[False-False-False-False-False] 51.6600μs 20.3733μs 49.0840 KOps/s 48.4982 KOps/s $\color{#35bf28}+1.21\%$
test_values[generalized_advantage_estimate-True-True] 25.1819ms 24.5715ms 40.6975 Ops/s 40.1740 Ops/s $\color{#35bf28}+1.30\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1012s 2.9176ms 342.7454 Ops/s 354.7170 Ops/s $\color{#d91a1a}-3.37\%$
test_values[td0_return_estimate-False-False] 0.1039ms 78.9108μs 12.6725 KOps/s 12.6776 KOps/s $\color{#d91a1a}-0.04\%$
test_values[td1_return_estimate-False-False] 55.9353ms 54.7062ms 18.2795 Ops/s 17.7737 Ops/s $\color{#35bf28}+2.85\%$
test_values[vec_td1_return_estimate-False-False] 1.3642ms 1.0765ms 928.9411 Ops/s 927.9410 Ops/s $\color{#35bf28}+0.11\%$
test_values[td_lambda_return_estimate-True-False] 91.2675ms 87.4780ms 11.4315 Ops/s 11.0831 Ops/s $\color{#35bf28}+3.14\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3826ms 1.0712ms 933.5326 Ops/s 928.9010 Ops/s $\color{#35bf28}+0.50\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.5991ms 24.3727ms 41.0295 Ops/s 41.0063 Ops/s $\color{#35bf28}+0.06\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0211ms 0.7426ms 1.3467 KOps/s 1.3440 KOps/s $\color{#35bf28}+0.20\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7663ms 0.6640ms 1.5060 KOps/s 1.5019 KOps/s $\color{#35bf28}+0.27\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5171ms 1.4698ms 680.3490 Ops/s 677.3657 Ops/s $\color{#35bf28}+0.44\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7235ms 0.6792ms 1.4724 KOps/s 1.4692 KOps/s $\color{#35bf28}+0.21\%$
test_dqn_speed[False-None] 1.6975ms 1.4851ms 673.3466 Ops/s 682.0454 Ops/s $\color{#d91a1a}-1.28\%$
test_dqn_speed[False-backward] 2.1350ms 2.0819ms 480.3263 Ops/s 484.9782 Ops/s $\color{#d91a1a}-0.96\%$
test_dqn_speed[True-None] 0.6008ms 0.5344ms 1.8711 KOps/s 1.8322 KOps/s $\color{#35bf28}+2.13\%$
test_dqn_speed[True-backward] 1.1572ms 1.0825ms 923.7671 Ops/s 829.7423 Ops/s $\textbf{\color{#35bf28}+11.33\%}$
test_dqn_speed[reduce-overhead-None] 0.6124ms 0.5534ms 1.8070 KOps/s 1.7796 KOps/s $\color{#35bf28}+1.54\%$
test_dqn_speed[reduce-overhead-backward] 0.9702ms 0.9331ms 1.0717 KOps/s 938.7859 Ops/s $\textbf{\color{#35bf28}+14.16\%}$
test_ddpg_speed[False-None] 3.1167ms 2.8159ms 355.1219 Ops/s 358.0406 Ops/s $\color{#d91a1a}-0.82\%$
test_ddpg_speed[False-backward] 4.5272ms 4.0825ms 244.9495 Ops/s 240.4532 Ops/s $\color{#35bf28}+1.87\%$
test_ddpg_speed[True-None] 1.1600ms 1.0577ms 945.4436 Ops/s 900.8113 Ops/s $\color{#35bf28}+4.95\%$
test_ddpg_speed[True-backward] 2.1822ms 2.0912ms 478.1880 Ops/s 435.2892 Ops/s $\textbf{\color{#35bf28}+9.86\%}$
test_ddpg_speed[reduce-overhead-None] 1.1572ms 1.0681ms 936.2624 Ops/s 916.6801 Ops/s $\color{#35bf28}+2.14\%$
test_ddpg_speed[reduce-overhead-backward] 1.6357ms 1.5912ms 628.4391 Ops/s 561.4132 Ops/s $\textbf{\color{#35bf28}+11.94\%}$
test_sac_speed[False-None] 8.3586ms 7.9346ms 126.0300 Ops/s 126.3663 Ops/s $\color{#d91a1a}-0.27\%$
test_sac_speed[False-backward] 11.3192ms 10.8092ms 92.5137 Ops/s 90.7442 Ops/s $\color{#35bf28}+1.95\%$
test_sac_speed[True-None] 1.5428ms 1.4930ms 669.8082 Ops/s 627.4888 Ops/s $\textbf{\color{#35bf28}+6.74\%}$
test_sac_speed[True-backward] 3.2537ms 3.1493ms 317.5275 Ops/s 296.5754 Ops/s $\textbf{\color{#35bf28}+7.06\%}$
test_sac_speed[reduce-overhead-None] 22.2010ms 12.3854ms 80.7403 Ops/s 79.6866 Ops/s $\color{#35bf28}+1.32\%$
test_sac_speed[reduce-overhead-backward] 1.3797ms 1.3140ms 761.0588 Ops/s 736.7991 Ops/s $\color{#35bf28}+3.29\%$
test_redq_speed[False-None] 8.1222ms 7.4044ms 135.0546 Ops/s 134.3991 Ops/s $\color{#35bf28}+0.49\%$
test_redq_speed[False-backward] 12.0974ms 11.1485ms 89.6982 Ops/s 89.0591 Ops/s $\color{#35bf28}+0.72\%$
test_redq_speed[True-None] 1.9823ms 1.9319ms 517.6261 Ops/s 509.5557 Ops/s $\color{#35bf28}+1.58\%$
test_redq_speed[True-backward] 3.6175ms 3.5239ms 283.7744 Ops/s 264.0639 Ops/s $\textbf{\color{#35bf28}+7.46\%}$
test_redq_speed[reduce-overhead-None] 2.0216ms 1.9452ms 514.0961 Ops/s 494.2823 Ops/s $\color{#35bf28}+4.01\%$
test_redq_speed[reduce-overhead-backward] 3.6335ms 3.5590ms 280.9814 Ops/s 263.8519 Ops/s $\textbf{\color{#35bf28}+6.49\%}$
test_redq_deprec_speed[False-None] 9.3466ms 8.9221ms 112.0809 Ops/s 111.2050 Ops/s $\color{#35bf28}+0.79\%$
test_redq_deprec_speed[False-backward] 12.3292ms 11.8705ms 84.2424 Ops/s 81.7790 Ops/s $\color{#35bf28}+3.01\%$
test_redq_deprec_speed[True-None] 2.3591ms 2.3035ms 434.1193 Ops/s 428.1359 Ops/s $\color{#35bf28}+1.40\%$
test_redq_deprec_speed[True-backward] 3.9715ms 3.8891ms 257.1317 Ops/s 244.9820 Ops/s $\color{#35bf28}+4.96\%$
test_redq_deprec_speed[reduce-overhead-None] 2.4035ms 2.2827ms 438.0738 Ops/s 432.1352 Ops/s $\color{#35bf28}+1.37\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.3016ms 3.9038ms 256.1599 Ops/s 243.5284 Ops/s $\textbf{\color{#35bf28}+5.19\%}$
test_td3_speed[False-None] 8.1191ms 7.8340ms 127.6487 Ops/s 128.0732 Ops/s $\color{#d91a1a}-0.33\%$
test_td3_speed[False-backward] 10.4973ms 10.0660ms 99.3445 Ops/s 96.8646 Ops/s $\color{#35bf28}+2.56\%$
test_td3_speed[True-None] 1.5866ms 1.5616ms 640.3733 Ops/s 630.7042 Ops/s $\color{#35bf28}+1.53\%$
test_td3_speed[True-backward] 3.1037ms 3.0453ms 328.3790 Ops/s 319.0149 Ops/s $\color{#35bf28}+2.94\%$
test_td3_speed[reduce-overhead-None] 48.7896ms 24.9514ms 40.0780 Ops/s 39.8103 Ops/s $\color{#35bf28}+0.67\%$
test_td3_speed[reduce-overhead-backward] 1.3457ms 1.2734ms 785.3256 Ops/s 778.4750 Ops/s $\color{#35bf28}+0.88\%$
test_cql_speed[False-None] 17.0777ms 16.5516ms 60.4173 Ops/s 60.2561 Ops/s $\color{#35bf28}+0.27\%$
test_cql_speed[False-backward] 22.3440ms 21.6322ms 46.2275 Ops/s 45.9170 Ops/s $\color{#35bf28}+0.68\%$
test_cql_speed[True-None] 2.9032ms 2.8594ms 349.7275 Ops/s 343.9967 Ops/s $\color{#35bf28}+1.67\%$
test_cql_speed[True-backward] 5.2664ms 5.0289ms 198.8501 Ops/s 193.7145 Ops/s $\color{#35bf28}+2.65\%$
test_cql_speed[reduce-overhead-None] 0.3590s 14.6558ms 68.2325 Ops/s 76.1097 Ops/s $\textbf{\color{#d91a1a}-10.35\%}$
test_cql_speed[reduce-overhead-backward] 1.5614ms 1.5014ms 666.0487 Ops/s 587.9584 Ops/s $\textbf{\color{#35bf28}+13.28\%}$
test_a2c_speed[False-None] 3.2591ms 3.1690ms 315.5589 Ops/s 309.0357 Ops/s $\color{#35bf28}+2.11\%$
test_a2c_speed[False-backward] 6.7101ms 6.0468ms 165.3776 Ops/s 156.9378 Ops/s $\textbf{\color{#35bf28}+5.38\%}$
test_a2c_speed[True-None] 1.0791ms 0.9896ms 1.0105 KOps/s 988.9371 Ops/s $\color{#35bf28}+2.18\%$
test_a2c_speed[True-backward] 2.5799ms 2.4994ms 400.0982 Ops/s 389.0158 Ops/s $\color{#35bf28}+2.85\%$
test_a2c_speed[reduce-overhead-None] 20.5522ms 11.1455ms 89.7227 Ops/s 89.4842 Ops/s $\color{#35bf28}+0.27\%$
test_a2c_speed[reduce-overhead-backward] 0.9916ms 0.9526ms 1.0497 KOps/s 876.6777 Ops/s $\textbf{\color{#35bf28}+19.74\%}$
test_ppo_speed[False-None] 3.7399ms 3.6402ms 274.7139 Ops/s 274.3704 Ops/s $\color{#35bf28}+0.13\%$
test_ppo_speed[False-backward] 7.1581ms 6.7227ms 148.7488 Ops/s 142.4868 Ops/s $\color{#35bf28}+4.39\%$
test_ppo_speed[True-None] 1.0483ms 0.9376ms 1.0666 KOps/s 1.0416 KOps/s $\color{#35bf28}+2.40\%$
test_ppo_speed[True-backward] 2.5625ms 2.4910ms 401.4510 Ops/s 400.4511 Ops/s $\color{#35bf28}+0.25\%$
test_ppo_speed[reduce-overhead-None] 0.5665ms 0.5183ms 1.9293 KOps/s 69.9518 Ops/s $\textbf{\color{#35bf28}+2658.06\%}$
test_ppo_speed[reduce-overhead-backward] 1.0032ms 0.9423ms 1.0612 KOps/s 996.3880 Ops/s $\textbf{\color{#35bf28}+6.51\%}$
test_reinforce_speed[False-None] 2.3676ms 2.2395ms 446.5379 Ops/s 449.4679 Ops/s $\color{#d91a1a}-0.65\%$
test_reinforce_speed[False-backward] 3.6791ms 3.2340ms 309.2145 Ops/s 310.4510 Ops/s $\color{#d91a1a}-0.40\%$
test_reinforce_speed[True-None] 0.8896ms 0.8120ms 1.2315 KOps/s 1.1478 KOps/s $\textbf{\color{#35bf28}+7.29\%}$
test_reinforce_speed[True-backward] 2.4392ms 2.3489ms 425.7310 Ops/s 415.0107 Ops/s $\color{#35bf28}+2.58\%$
test_reinforce_speed[reduce-overhead-None] 0.2937s 11.7526ms 85.0877 Ops/s 93.0136 Ops/s $\textbf{\color{#d91a1a}-8.52\%}$
test_reinforce_speed[reduce-overhead-backward] 1.0612ms 1.0138ms 986.3522 Ops/s 943.8642 Ops/s $\color{#35bf28}+4.50\%$
test_iql_speed[False-None] 10.1593ms 9.2946ms 107.5895 Ops/s 109.3238 Ops/s $\color{#d91a1a}-1.59\%$
test_iql_speed[False-backward] 13.5316ms 12.9220ms 77.3873 Ops/s 77.6809 Ops/s $\color{#d91a1a}-0.38\%$
test_iql_speed[True-None] 1.7932ms 1.7304ms 577.8925 Ops/s 573.0355 Ops/s $\color{#35bf28}+0.85\%$
test_iql_speed[True-backward] 4.4554ms 4.3113ms 231.9490 Ops/s 229.8969 Ops/s $\color{#35bf28}+0.89\%$
test_iql_speed[reduce-overhead-None] 19.7536ms 11.1602ms 89.6043 Ops/s 70.5586 Ops/s $\textbf{\color{#35bf28}+26.99\%}$
test_iql_speed[reduce-overhead-backward] 1.6397ms 1.5688ms 637.4375 Ops/s 622.2715 Ops/s $\color{#35bf28}+2.44\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7657ms 6.1954ms 161.4111 Ops/s 159.1675 Ops/s $\color{#35bf28}+1.41\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5090ms 0.2728ms 3.6654 KOps/s 3.6292 KOps/s $\color{#35bf28}+1.00\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4318ms 0.2628ms 3.8045 KOps/s 3.9580 KOps/s $\color{#d91a1a}-3.88\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2045ms 5.9293ms 168.6553 Ops/s 166.8643 Ops/s $\color{#35bf28}+1.07\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0957ms 0.2912ms 3.4346 KOps/s 3.8220 KOps/s $\textbf{\color{#d91a1a}-10.14\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5045ms 0.2875ms 3.4788 KOps/s 2.8193 KOps/s $\textbf{\color{#35bf28}+23.39\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6344ms 1.4465ms 691.3473 Ops/s 806.8459 Ops/s $\textbf{\color{#d91a1a}-14.31\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.9653ms 1.3475ms 742.1206 Ops/s 880.1377 Ops/s $\textbf{\color{#d91a1a}-15.68\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2530ms 6.1160ms 163.5044 Ops/s 161.1503 Ops/s $\color{#35bf28}+1.46\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8901ms 0.4253ms 2.3511 KOps/s 2.1767 KOps/s $\textbf{\color{#35bf28}+8.01\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6127ms 0.3913ms 2.5555 KOps/s 2.2745 KOps/s $\textbf{\color{#35bf28}+12.36\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0173ms 5.9093ms 169.2239 Ops/s 166.8858 Ops/s $\color{#35bf28}+1.40\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7679ms 0.3495ms 2.8609 KOps/s 2.9028 KOps/s $\color{#d91a1a}-1.44\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5194ms 0.3233ms 3.0935 KOps/s 3.4811 KOps/s $\textbf{\color{#d91a1a}-11.13\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1216ms 5.8740ms 170.2406 Ops/s 168.2699 Ops/s $\color{#35bf28}+1.17\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.5721ms 0.3216ms 3.1095 KOps/s 3.2859 KOps/s $\textbf{\color{#d91a1a}-5.37\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5059ms 0.2757ms 3.6277 KOps/s 3.3334 KOps/s $\textbf{\color{#35bf28}+8.83\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1755ms 6.0008ms 166.6431 Ops/s 162.9003 Ops/s $\color{#35bf28}+2.30\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8182ms 0.4584ms 2.1817 KOps/s 2.3980 KOps/s $\textbf{\color{#d91a1a}-9.02\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6529ms 0.4372ms 2.2874 KOps/s 2.2355 KOps/s $\color{#35bf28}+2.32\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0851ms 5.4092ms 184.8719 Ops/s 188.9140 Ops/s $\color{#d91a1a}-2.14\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.8143ms 1.9335ms 517.1842 Ops/s 452.1710 Ops/s $\textbf{\color{#35bf28}+14.38\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.9263ms 1.2056ms 829.4930 Ops/s 809.4886 Ops/s $\color{#35bf28}+2.47\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.3392ms 5.3171ms 188.0717 Ops/s 188.7334 Ops/s $\color{#d91a1a}-0.35\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.3804ms 2.0091ms 497.7328 Ops/s 505.4047 Ops/s $\color{#d91a1a}-1.52\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.9778ms 1.2132ms 824.2850 Ops/s 862.2730 Ops/s $\color{#d91a1a}-4.41\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4948s 15.4273ms 64.8200 Ops/s 33.0584 Ops/s $\textbf{\color{#35bf28}+96.08\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.8573ms 2.2274ms 448.9512 Ops/s 449.1083 Ops/s $\color{#d91a1a}-0.03\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.2806ms 1.3113ms 762.5833 Ops/s 738.6237 Ops/s $\color{#35bf28}+3.24\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 17.7752ms 15.8691ms 63.0155 Ops/s 66.4099 Ops/s $\textbf{\color{#d91a1a}-5.11\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.8000ms 17.4270ms 57.3824 Ops/s 57.7434 Ops/s $\color{#d91a1a}-0.63\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 20.5297ms 19.8756ms 50.3129 Ops/s 50.6371 Ops/s $\color{#d91a1a}-0.64\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.1519ms 17.5687ms 56.9194 Ops/s 57.5214 Ops/s $\color{#d91a1a}-1.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.2792ms 19.7282ms 50.6887 Ops/s 51.3534 Ops/s $\color{#d91a1a}-1.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.4172ms 19.4847ms 51.3223 Ops/s 52.4788 Ops/s $\color{#d91a1a}-2.20\%$

@vmoens vmoens closed this Jan 10, 2025
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