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

[Doc] Better doc for inverse transform semantic #2459

Open
wants to merge 1 commit into
base: gh/vmoens/30/base
Choose a base branch
from

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Sep 30, 2024

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Sep 30, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2459

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:

❌ 4 New Failures, 6 Unrelated Failures

As of commit 679a3ae with merge base a0dfddc (image):

NEW FAILURES - The following jobs have failed:

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 Sep 30, 2024
ghstack-source-id: 444ad87d1ab0a829e8ce1848b5838859d5ee7494
Pull Request resolved: #2459
@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 Sep 30, 2024
Copy link

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

Total Benchmarks: 146. Improved: $\large\color{#35bf28}39$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 62.9869ms 60.5555ms 16.5138 Ops/s 16.6085 Ops/s $\color{#d91a1a}-0.57\%$
test_sync 39.2449ms 34.0208ms 29.3938 Ops/s 30.7106 Ops/s $\color{#d91a1a}-4.29\%$
test_async 0.1524s 31.4843ms 31.7619 Ops/s 31.8682 Ops/s $\color{#d91a1a}-0.33\%$
test_simple 0.5261s 0.4375s 2.2856 Ops/s 2.4670 Ops/s $\textbf{\color{#d91a1a}-7.35\%}$
test_transformed 0.6759s 0.5933s 1.6854 Ops/s 1.6945 Ops/s $\color{#d91a1a}-0.53\%$
test_serial 1.3817s 1.3019s 0.7681 Ops/s 0.7704 Ops/s $\color{#d91a1a}-0.29\%$
test_parallel 1.2414s 1.1515s 0.8685 Ops/s 0.8661 Ops/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-True-True-True-True] 0.2517ms 27.2664μs 36.6752 KOps/s 35.9239 KOps/s $\color{#35bf28}+2.09\%$
test_step_mdp_speed[True-True-True-True-False] 64.0200μs 15.8320μs 63.1634 KOps/s 60.9797 KOps/s $\color{#35bf28}+3.58\%$
test_step_mdp_speed[True-True-True-False-True] 45.9160μs 15.4394μs 64.7692 KOps/s 63.0115 KOps/s $\color{#35bf28}+2.79\%$
test_step_mdp_speed[True-True-True-False-False] 53.2500μs 9.1758μs 108.9824 KOps/s 106.5182 KOps/s $\color{#35bf28}+2.31\%$
test_step_mdp_speed[True-True-False-True-True] 75.5120μs 29.1015μs 34.3624 KOps/s 34.1122 KOps/s $\color{#35bf28}+0.73\%$
test_step_mdp_speed[True-True-False-True-False] 57.9680μs 17.5458μs 56.9936 KOps/s 55.3696 KOps/s $\color{#35bf28}+2.93\%$
test_step_mdp_speed[True-True-False-False-True] 46.4670μs 17.2249μs 58.0555 KOps/s 56.2950 KOps/s $\color{#35bf28}+3.13\%$
test_step_mdp_speed[True-True-False-False-False] 62.4570μs 10.7740μs 92.8158 KOps/s 90.1529 KOps/s $\color{#35bf28}+2.95\%$
test_step_mdp_speed[True-False-True-True-True] 79.9090μs 30.7800μs 32.4887 KOps/s 31.9289 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[True-False-True-True-False] 70.2010μs 19.2472μs 51.9556 KOps/s 50.5671 KOps/s $\color{#35bf28}+2.75\%$
test_step_mdp_speed[True-False-True-False-True] 62.0960μs 17.2498μs 57.9718 KOps/s 56.1332 KOps/s $\color{#35bf28}+3.28\%$
test_step_mdp_speed[True-False-True-False-False] 66.3440μs 10.7644μs 92.8987 KOps/s 89.1058 KOps/s $\color{#35bf28}+4.26\%$
test_step_mdp_speed[True-False-False-True-True] 85.4700μs 32.1793μs 31.0758 KOps/s 30.5874 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[True-False-False-True-False] 58.3290μs 20.6527μs 48.4199 KOps/s 47.3340 KOps/s $\color{#35bf28}+2.29\%$
test_step_mdp_speed[True-False-False-False-True] 86.6320μs 18.7389μs 53.3650 KOps/s 52.4274 KOps/s $\color{#35bf28}+1.79\%$
test_step_mdp_speed[True-False-False-False-False] 42.3400μs 12.3919μs 80.6982 KOps/s 78.9355 KOps/s $\color{#35bf28}+2.23\%$
test_step_mdp_speed[False-True-True-True-True] 87.1720μs 30.6683μs 32.6069 KOps/s 31.9997 KOps/s $\color{#35bf28}+1.90\%$
test_step_mdp_speed[False-True-True-True-False] 59.7320μs 19.4786μs 51.3383 KOps/s 51.0027 KOps/s $\color{#35bf28}+0.66\%$
test_step_mdp_speed[False-True-True-False-True] 75.5520μs 19.3400μs 51.7063 KOps/s 50.4355 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[False-True-True-False-False] 47.0380μs 12.1229μs 82.4886 KOps/s 81.2018 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[False-True-False-True-True] 90.2990μs 31.8365μs 31.4105 KOps/s 30.3289 KOps/s $\color{#35bf28}+3.57\%$
test_step_mdp_speed[False-True-False-True-False] 75.7420μs 20.7952μs 48.0881 KOps/s 47.3569 KOps/s $\color{#35bf28}+1.54\%$
test_step_mdp_speed[False-True-False-False-True] 3.0593ms 21.4109μs 46.7053 KOps/s 45.7128 KOps/s $\color{#35bf28}+2.17\%$
test_step_mdp_speed[False-True-False-False-False] 65.1610μs 13.5271μs 73.9259 KOps/s 72.2624 KOps/s $\color{#35bf28}+2.30\%$
test_step_mdp_speed[False-False-True-True-True] 79.8090μs 33.4860μs 29.8632 KOps/s 29.1064 KOps/s $\color{#35bf28}+2.60\%$
test_step_mdp_speed[False-False-True-True-False] 75.0000μs 22.1146μs 45.2190 KOps/s 43.9812 KOps/s $\color{#35bf28}+2.81\%$
test_step_mdp_speed[False-False-True-False-True] 74.4390μs 21.3597μs 46.8171 KOps/s 46.1308 KOps/s $\color{#35bf28}+1.49\%$
test_step_mdp_speed[False-False-True-False-False] 42.8100μs 13.4710μs 74.2338 KOps/s 72.8520 KOps/s $\color{#35bf28}+1.90\%$
test_step_mdp_speed[False-False-False-True-True] 82.6340μs 35.2753μs 28.3484 KOps/s 28.3895 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[False-False-False-True-False] 52.6180μs 23.5644μs 42.4369 KOps/s 41.1701 KOps/s $\color{#35bf28}+3.08\%$
test_step_mdp_speed[False-False-False-False-True] 71.2830μs 22.2126μs 45.0195 KOps/s 43.8954 KOps/s $\color{#35bf28}+2.56\%$
test_step_mdp_speed[False-False-False-False-False] 42.0480μs 15.0116μs 66.6152 KOps/s 66.5084 KOps/s $\color{#35bf28}+0.16\%$
test_values[generalized_advantage_estimate-True-True] 9.9503ms 9.4394ms 105.9392 Ops/s 105.3867 Ops/s $\color{#35bf28}+0.52\%$
test_values[vec_generalized_advantage_estimate-True-True] 40.5085ms 33.4798ms 29.8688 Ops/s 29.7359 Ops/s $\color{#35bf28}+0.45\%$
test_values[td0_return_estimate-False-False] 0.2140ms 0.1660ms 6.0245 KOps/s 5.6851 KOps/s $\textbf{\color{#35bf28}+5.97\%}$
test_values[td1_return_estimate-False-False] 27.4383ms 23.5867ms 42.3968 Ops/s 42.8576 Ops/s $\color{#d91a1a}-1.08\%$
test_values[vec_td1_return_estimate-False-False] 35.6115ms 33.3390ms 29.9949 Ops/s 29.6223 Ops/s $\color{#35bf28}+1.26\%$
test_values[td_lambda_return_estimate-True-False] 34.6208ms 34.1279ms 29.3016 Ops/s 29.6524 Ops/s $\color{#d91a1a}-1.18\%$
test_values[vec_td_lambda_return_estimate-True-False] 35.2290ms 33.3911ms 29.9481 Ops/s 29.6684 Ops/s $\color{#35bf28}+0.94\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.5421ms 8.2685ms 120.9406 Ops/s 120.2147 Ops/s $\color{#35bf28}+0.60\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3935ms 2.0185ms 495.4057 Ops/s 521.6765 Ops/s $\textbf{\color{#d91a1a}-5.04\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5164ms 0.3630ms 2.7551 KOps/s 2.7500 KOps/s $\color{#35bf28}+0.18\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 47.5724ms 44.5517ms 22.4458 Ops/s 24.9565 Ops/s $\textbf{\color{#d91a1a}-10.06\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.9403ms 3.0239ms 330.7007 Ops/s 328.7274 Ops/s $\color{#35bf28}+0.60\%$
test_dqn_speed[False-None] 5.7809ms 1.3549ms 738.0846 Ops/s 732.0489 Ops/s $\color{#35bf28}+0.82\%$
test_dqn_speed[False-backward] 2.0258ms 1.8728ms 533.9617 Ops/s 538.9601 Ops/s $\color{#d91a1a}-0.93\%$
test_dqn_speed[True-None] 0.5796ms 0.4645ms 2.1529 KOps/s 2.0935 KOps/s $\color{#35bf28}+2.83\%$
test_dqn_speed[True-backward] 1.0708ms 0.9472ms 1.0557 KOps/s 1.1128 KOps/s $\textbf{\color{#d91a1a}-5.13\%}$
test_dqn_speed[reduce-overhead-None] 0.5741ms 0.4657ms 2.1473 KOps/s 2.0826 KOps/s $\color{#35bf28}+3.11\%$
test_dqn_speed[reduce-overhead-backward] 1.0568ms 0.9349ms 1.0696 KOps/s 1.0554 KOps/s $\color{#35bf28}+1.35\%$
test_ddpg_speed[False-None] 3.7560ms 2.8134ms 355.4373 Ops/s 350.3456 Ops/s $\color{#35bf28}+1.45\%$
test_ddpg_speed[False-backward] 6.0223ms 4.0628ms 246.1358 Ops/s 248.3636 Ops/s $\color{#d91a1a}-0.90\%$
test_ddpg_speed[True-None] 1.2641ms 1.0166ms 983.6854 Ops/s 949.3166 Ops/s $\color{#35bf28}+3.62\%$
test_ddpg_speed[True-backward] 2.2435ms 1.9486ms 513.1926 Ops/s 492.1142 Ops/s $\color{#35bf28}+4.28\%$
test_ddpg_speed[reduce-overhead-None] 1.4170ms 1.0170ms 983.2570 Ops/s 932.3969 Ops/s $\textbf{\color{#35bf28}+5.45\%}$
test_ddpg_speed[reduce-overhead-backward] 2.3143ms 1.9797ms 505.1195 Ops/s 499.9036 Ops/s $\color{#35bf28}+1.04\%$
test_sac_speed[False-None] 12.9634ms 8.1406ms 122.8416 Ops/s 94.1094 Ops/s $\textbf{\color{#35bf28}+30.53\%}$
test_sac_speed[False-backward] 11.7077ms 10.9526ms 91.3025 Ops/s 87.3541 Ops/s $\color{#35bf28}+4.52\%$
test_sac_speed[True-None] 2.3890ms 1.8592ms 537.8558 Ops/s 513.8751 Ops/s $\color{#35bf28}+4.67\%$
test_sac_speed[True-backward] 4.2249ms 3.6138ms 276.7182 Ops/s 262.5990 Ops/s $\textbf{\color{#35bf28}+5.38\%}$
test_sac_speed[reduce-overhead-None] 2.8158ms 1.8736ms 533.7257 Ops/s 513.4508 Ops/s $\color{#35bf28}+3.95\%$
test_sac_speed[reduce-overhead-backward] 4.1669ms 3.6307ms 275.4269 Ops/s 267.4394 Ops/s $\color{#35bf28}+2.99\%$
test_redq_speed[False-None] 15.1430ms 13.0045ms 76.8965 Ops/s 74.0335 Ops/s $\color{#35bf28}+3.87\%$
test_redq_speed[False-backward] 44.9987ms 22.9009ms 43.6664 Ops/s 43.6133 Ops/s $\color{#35bf28}+0.12\%$
test_redq_speed[True-None] 7.2826ms 4.6966ms 212.9202 Ops/s 189.3514 Ops/s $\textbf{\color{#35bf28}+12.45\%}$
test_redq_speed[True-backward] 13.2788ms 12.4227ms 80.4976 Ops/s 76.1178 Ops/s $\textbf{\color{#35bf28}+5.75\%}$
test_redq_speed[reduce-overhead-None] 5.7076ms 4.4943ms 222.5038 Ops/s 191.1318 Ops/s $\textbf{\color{#35bf28}+16.41\%}$
test_redq_speed[reduce-overhead-backward] 12.7946ms 12.0723ms 82.8345 Ops/s 77.3444 Ops/s $\textbf{\color{#35bf28}+7.10\%}$
test_redq_deprec_speed[False-None] 14.2862ms 12.9502ms 77.2187 Ops/s 72.5565 Ops/s $\textbf{\color{#35bf28}+6.43\%}$
test_redq_deprec_speed[False-backward] 21.3321ms 18.8798ms 52.9666 Ops/s 49.4792 Ops/s $\textbf{\color{#35bf28}+7.05\%}$
test_redq_deprec_speed[True-None] 4.2031ms 3.5579ms 281.0636 Ops/s 263.5477 Ops/s $\textbf{\color{#35bf28}+6.65\%}$
test_redq_deprec_speed[True-backward] 8.3711ms 7.9037ms 126.5228 Ops/s 112.9420 Ops/s $\textbf{\color{#35bf28}+12.02\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.1517ms 3.5512ms 281.5916 Ops/s 223.7918 Ops/s $\textbf{\color{#35bf28}+25.83\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.2831ms 8.5304ms 117.2279 Ops/s 111.9895 Ops/s $\color{#35bf28}+4.68\%$
test_td3_speed[False-None] 8.3580ms 7.9831ms 125.2643 Ops/s 117.4301 Ops/s $\textbf{\color{#35bf28}+6.67\%}$
test_td3_speed[False-backward] 15.6373ms 10.8485ms 92.1784 Ops/s 80.9184 Ops/s $\textbf{\color{#35bf28}+13.92\%}$
test_td3_speed[True-None] 2.1945ms 1.9328ms 517.3815 Ops/s 483.5382 Ops/s $\textbf{\color{#35bf28}+7.00\%}$
test_td3_speed[True-backward] 3.9042ms 3.5951ms 278.1581 Ops/s 262.5855 Ops/s $\textbf{\color{#35bf28}+5.93\%}$
test_td3_speed[reduce-overhead-None] 2.2113ms 1.9091ms 523.8106 Ops/s 492.4993 Ops/s $\textbf{\color{#35bf28}+6.36\%}$
test_td3_speed[reduce-overhead-backward] 5.6694ms 3.7021ms 270.1161 Ops/s 260.3028 Ops/s $\color{#35bf28}+3.77\%$
test_cql_speed[False-None] 38.4139ms 35.7220ms 27.9939 Ops/s 26.9071 Ops/s $\color{#35bf28}+4.04\%$
test_cql_speed[False-backward] 47.7352ms 45.3919ms 22.0304 Ops/s 20.6345 Ops/s $\textbf{\color{#35bf28}+6.76\%}$
test_cql_speed[True-None] 16.9699ms 15.6215ms 64.0143 Ops/s 60.2390 Ops/s $\textbf{\color{#35bf28}+6.27\%}$
test_cql_speed[True-backward] 24.5033ms 22.6894ms 44.0734 Ops/s 42.0744 Ops/s $\color{#35bf28}+4.75\%$
test_cql_speed[reduce-overhead-None] 16.8943ms 15.8441ms 63.1151 Ops/s 61.2166 Ops/s $\color{#35bf28}+3.10\%$
test_cql_speed[reduce-overhead-backward] 23.4628ms 22.2103ms 45.0242 Ops/s 42.5453 Ops/s $\textbf{\color{#35bf28}+5.83\%}$
test_a2c_speed[False-None] 0.3749s 10.8095ms 92.5115 Ops/s 125.2140 Ops/s $\textbf{\color{#d91a1a}-26.12\%}$
test_a2c_speed[False-backward] 16.4091ms 14.7896ms 67.6152 Ops/s 64.4632 Ops/s $\color{#35bf28}+4.89\%$
test_a2c_speed[True-None] 4.1158ms 3.3246ms 300.7904 Ops/s 294.0573 Ops/s $\color{#35bf28}+2.29\%$
test_a2c_speed[True-backward] 11.0888ms 10.3740ms 96.3952 Ops/s 90.0052 Ops/s $\textbf{\color{#35bf28}+7.10\%}$
test_a2c_speed[reduce-overhead-None] 7.6130ms 3.3989ms 294.2119 Ops/s 293.8291 Ops/s $\color{#35bf28}+0.13\%$
test_a2c_speed[reduce-overhead-backward] 10.8171ms 10.1435ms 98.5853 Ops/s 92.9556 Ops/s $\textbf{\color{#35bf28}+6.06\%}$
test_ppo_speed[False-None] 8.8115ms 7.6224ms 131.1931 Ops/s 124.5173 Ops/s $\textbf{\color{#35bf28}+5.36\%}$
test_ppo_speed[False-backward] 17.0371ms 15.2002ms 65.7888 Ops/s 63.5829 Ops/s $\color{#35bf28}+3.47\%$
test_ppo_speed[True-None] 4.2703ms 3.7495ms 266.7010 Ops/s 256.5891 Ops/s $\color{#35bf28}+3.94\%$
test_ppo_speed[True-backward] 11.5645ms 9.9832ms 100.1679 Ops/s 99.1004 Ops/s $\color{#35bf28}+1.08\%$
test_ppo_speed[reduce-overhead-None] 4.3126ms 3.8311ms 261.0245 Ops/s 259.4575 Ops/s $\color{#35bf28}+0.60\%$
test_ppo_speed[reduce-overhead-backward] 11.1170ms 9.7986ms 102.0555 Ops/s 98.8879 Ops/s $\color{#35bf28}+3.20\%$
test_reinforce_speed[False-None] 7.8123ms 6.5139ms 153.5190 Ops/s 150.7600 Ops/s $\color{#35bf28}+1.83\%$
test_reinforce_speed[False-backward] 12.5173ms 9.9587ms 100.4152 Ops/s 98.7046 Ops/s $\color{#35bf28}+1.73\%$
test_reinforce_speed[True-None] 3.3149ms 2.6496ms 377.4147 Ops/s 369.5529 Ops/s $\color{#35bf28}+2.13\%$
test_reinforce_speed[True-backward] 9.4796ms 8.6492ms 115.6180 Ops/s 109.3584 Ops/s $\textbf{\color{#35bf28}+5.72\%}$
test_reinforce_speed[reduce-overhead-None] 3.2830ms 2.6586ms 376.1326 Ops/s 366.7796 Ops/s $\color{#35bf28}+2.55\%$
test_reinforce_speed[reduce-overhead-backward] 9.4908ms 8.9489ms 111.7452 Ops/s 111.1828 Ops/s $\color{#35bf28}+0.51\%$
test_iql_speed[False-None] 33.8073ms 32.4716ms 30.7962 Ops/s 30.6088 Ops/s $\color{#35bf28}+0.61\%$
test_iql_speed[False-backward] 47.9309ms 45.4480ms 22.0031 Ops/s 21.7691 Ops/s $\color{#35bf28}+1.08\%$
test_iql_speed[True-None] 14.4212ms 13.5578ms 73.7582 Ops/s 69.3656 Ops/s $\textbf{\color{#35bf28}+6.33\%}$
test_iql_speed[True-backward] 26.1530ms 24.8462ms 40.2477 Ops/s 38.6406 Ops/s $\color{#35bf28}+4.16\%$
test_iql_speed[reduce-overhead-None] 14.5277ms 13.4173ms 74.5307 Ops/s 70.2970 Ops/s $\textbf{\color{#35bf28}+6.02\%}$
test_iql_speed[reduce-overhead-backward] 25.7142ms 24.7259ms 40.4434 Ops/s 38.3494 Ops/s $\textbf{\color{#35bf28}+5.46\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.4771ms 5.2452ms 190.6512 Ops/s 178.4080 Ops/s $\textbf{\color{#35bf28}+6.86\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0354ms 0.4860ms 2.0576 KOps/s 1.9165 KOps/s $\textbf{\color{#35bf28}+7.37\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7985ms 0.4592ms 2.1775 KOps/s 2.1015 KOps/s $\color{#35bf28}+3.62\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0604ms 5.1868ms 192.7989 Ops/s 184.8258 Ops/s $\color{#35bf28}+4.31\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0222ms 0.4812ms 2.0783 KOps/s 2.0241 KOps/s $\color{#35bf28}+2.68\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6812ms 0.4477ms 2.2337 KOps/s 2.1527 KOps/s $\color{#35bf28}+3.76\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.4891ms 1.6016ms 624.3673 Ops/s 618.6826 Ops/s $\color{#35bf28}+0.92\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.8467ms 1.5635ms 639.5717 Ops/s 640.7267 Ops/s $\color{#d91a1a}-0.18\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 8.3590ms 5.3201ms 187.9674 Ops/s 173.8719 Ops/s $\textbf{\color{#35bf28}+8.11\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0006ms 0.6246ms 1.6009 KOps/s 1.5480 KOps/s $\color{#35bf28}+3.42\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9571ms 0.5915ms 1.6908 KOps/s 1.6202 KOps/s $\color{#35bf28}+4.36\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.5870ms 5.2066ms 192.0637 Ops/s 179.3698 Ops/s $\textbf{\color{#35bf28}+7.08\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0648ms 0.4859ms 2.0582 KOps/s 1.9798 KOps/s $\color{#35bf28}+3.96\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6111ms 0.4519ms 2.2128 KOps/s 2.1021 KOps/s $\textbf{\color{#35bf28}+5.27\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.7261ms 5.1370ms 194.6675 Ops/s 181.2320 Ops/s $\textbf{\color{#35bf28}+7.41\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.4428s 1.0618ms 941.8217 Ops/s 2.0156 KOps/s $\textbf{\color{#d91a1a}-53.27\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6583ms 0.4516ms 2.2146 KOps/s 2.1064 KOps/s $\textbf{\color{#35bf28}+5.14\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.4234ms 5.3073ms 188.4214 Ops/s 176.4879 Ops/s $\textbf{\color{#35bf28}+6.76\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.7037ms 0.6263ms 1.5966 KOps/s 1.5497 KOps/s $\color{#35bf28}+3.03\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7797ms 0.5889ms 1.6980 KOps/s 1.5038 KOps/s $\textbf{\color{#35bf28}+12.92\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.3506ms 4.2690ms 234.2489 Ops/s 221.2652 Ops/s $\textbf{\color{#35bf28}+5.87\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.4657ms 2.2630ms 441.8928 Ops/s 442.4026 Ops/s $\color{#d91a1a}-0.12\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.2148ms 1.4052ms 711.6257 Ops/s 699.1395 Ops/s $\color{#35bf28}+1.79\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.3731s 11.6093ms 86.1377 Ops/s 235.0959 Ops/s $\textbf{\color{#d91a1a}-63.36\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.1470ms 1.9866ms 503.3621 Ops/s 427.8421 Ops/s $\textbf{\color{#35bf28}+17.65\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.6648ms 1.3461ms 742.8699 Ops/s 725.9186 Ops/s $\color{#35bf28}+2.34\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.8391ms 4.3975ms 227.4006 Ops/s 227.5464 Ops/s $\color{#d91a1a}-0.06\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.3997ms 2.5056ms 399.1087 Ops/s 477.7035 Ops/s $\textbf{\color{#d91a1a}-16.45\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.6185ms 1.4642ms 682.9570 Ops/s 706.1840 Ops/s $\color{#d91a1a}-3.29\%$

Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1039s 0.1034s 9.6728 Ops/s 9.7099 Ops/s $\color{#d91a1a}-0.38\%$
test_sync 93.5527ms 89.3832ms 11.1878 Ops/s 10.9729 Ops/s $\color{#35bf28}+1.96\%$
test_async 0.1491s 85.9113ms 11.6399 Ops/s 11.5392 Ops/s $\color{#35bf28}+0.87\%$
test_single_pixels 0.1097s 0.1095s 9.1354 Ops/s 9.1419 Ops/s $\color{#d91a1a}-0.07\%$
test_sync_pixels 72.6714ms 71.4422ms 13.9973 Ops/s 14.0093 Ops/s $\color{#d91a1a}-0.09\%$
test_async_pixels 0.1264s 67.2864ms 14.8618 Ops/s 14.9758 Ops/s $\color{#d91a1a}-0.76\%$
test_simple 0.7555s 0.7523s 1.3293 Ops/s 1.3284 Ops/s $\color{#35bf28}+0.07\%$
test_transformed 0.9775s 0.9687s 1.0323 Ops/s 1.0436 Ops/s $\color{#d91a1a}-1.08\%$
test_serial 2.2177s 2.1057s 0.4749 Ops/s 0.4788 Ops/s $\color{#d91a1a}-0.81\%$
test_parallel 1.9203s 1.9029s 0.5255 Ops/s 0.5333 Ops/s $\color{#d91a1a}-1.45\%$
test_step_mdp_speed[True-True-True-True-True] 0.2535ms 37.5118μs 26.6583 KOps/s 27.0158 KOps/s $\color{#d91a1a}-1.32\%$
test_step_mdp_speed[True-True-True-True-False] 46.8910μs 21.2855μs 46.9804 KOps/s 46.6943 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[True-True-True-False-True] 51.5510μs 21.6357μs 46.2198 KOps/s 46.3958 KOps/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[True-True-True-False-False] 45.1710μs 12.1367μs 82.3946 KOps/s 81.9025 KOps/s $\color{#35bf28}+0.60\%$
test_step_mdp_speed[True-True-False-True-True] 69.9200μs 39.2635μs 25.4690 KOps/s 25.1671 KOps/s $\color{#35bf28}+1.20\%$
test_step_mdp_speed[True-True-False-True-False] 54.5710μs 23.3801μs 42.7714 KOps/s 42.2051 KOps/s $\color{#35bf28}+1.34\%$
test_step_mdp_speed[True-True-False-False-True] 57.1610μs 22.7411μs 43.9732 KOps/s 42.3761 KOps/s $\color{#35bf28}+3.77\%$
test_step_mdp_speed[True-True-False-False-False] 44.2600μs 14.0460μs 71.1946 KOps/s 69.5166 KOps/s $\color{#35bf28}+2.41\%$
test_step_mdp_speed[True-False-True-True-True] 77.0310μs 41.8674μs 23.8849 KOps/s 23.6888 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[True-False-True-True-False] 53.9310μs 25.0901μs 39.8563 KOps/s 38.9611 KOps/s $\color{#35bf28}+2.30\%$
test_step_mdp_speed[True-False-True-False-True] 59.4010μs 23.2800μs 42.9554 KOps/s 42.7241 KOps/s $\color{#35bf28}+0.54\%$
test_step_mdp_speed[True-False-True-False-False] 46.3800μs 14.3853μs 69.5154 KOps/s 69.7221 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-False-False-True-True] 88.1910μs 43.9172μs 22.7701 KOps/s 22.5851 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-False-False-True-False] 62.4710μs 27.5292μs 36.3251 KOps/s 36.3917 KOps/s $\color{#d91a1a}-0.18\%$
test_step_mdp_speed[True-False-False-False-True] 54.4210μs 24.8407μs 40.2565 KOps/s 39.2756 KOps/s $\color{#35bf28}+2.50\%$
test_step_mdp_speed[True-False-False-False-False] 51.4010μs 16.1621μs 61.8732 KOps/s 61.8203 KOps/s $\color{#35bf28}+0.09\%$
test_step_mdp_speed[False-True-True-True-True] 74.5110μs 41.8665μs 23.8854 KOps/s 23.7358 KOps/s $\color{#35bf28}+0.63\%$
test_step_mdp_speed[False-True-True-True-False] 62.5010μs 25.2237μs 39.6452 KOps/s 39.2060 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[False-True-True-False-True] 62.7910μs 26.2230μs 38.1345 KOps/s 37.1690 KOps/s $\color{#35bf28}+2.60\%$
test_step_mdp_speed[False-True-True-False-False] 50.5300μs 15.7467μs 63.5054 KOps/s 63.0508 KOps/s $\color{#35bf28}+0.72\%$
test_step_mdp_speed[False-True-False-True-True] 73.6110μs 43.3427μs 23.0719 KOps/s 22.8065 KOps/s $\color{#35bf28}+1.16\%$
test_step_mdp_speed[False-True-False-True-False] 53.6400μs 27.1587μs 36.8207 KOps/s 36.1389 KOps/s $\color{#35bf28}+1.89\%$
test_step_mdp_speed[False-True-False-False-True] 3.6216ms 28.6824μs 34.8646 KOps/s 34.2682 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[False-True-False-False-False] 50.0410μs 17.7649μs 56.2908 KOps/s 56.4982 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[False-False-True-True-True] 88.5010μs 45.9350μs 21.7699 KOps/s 21.6476 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-False-True-True-False] 64.2210μs 29.5602μs 33.8293 KOps/s 33.4607 KOps/s $\color{#35bf28}+1.10\%$
test_step_mdp_speed[False-False-True-False-True] 57.3510μs 28.1751μs 35.4923 KOps/s 34.9230 KOps/s $\color{#35bf28}+1.63\%$
test_step_mdp_speed[False-False-True-False-False] 47.1110μs 17.6107μs 56.7837 KOps/s 55.9198 KOps/s $\color{#35bf28}+1.54\%$
test_step_mdp_speed[False-False-False-True-True] 81.8710μs 46.8587μs 21.3407 KOps/s 21.0609 KOps/s $\color{#35bf28}+1.33\%$
test_step_mdp_speed[False-False-False-True-False] 63.6110μs 31.1164μs 32.1374 KOps/s 30.8842 KOps/s $\color{#35bf28}+4.06\%$
test_step_mdp_speed[False-False-False-False-True] 66.7610μs 29.2353μs 34.2052 KOps/s 33.0466 KOps/s $\color{#35bf28}+3.51\%$
test_step_mdp_speed[False-False-False-False-False] 48.6900μs 19.3014μs 51.8098 KOps/s 50.3391 KOps/s $\color{#35bf28}+2.92\%$
test_values[generalized_advantage_estimate-True-True] 25.7440ms 25.1299ms 39.7933 Ops/s 40.3804 Ops/s $\color{#d91a1a}-1.45\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1201s 3.2866ms 304.2643 Ops/s 312.9004 Ops/s $\color{#d91a1a}-2.76\%$
test_values[td0_return_estimate-False-False] 90.0710μs 66.4652μs 15.0455 KOps/s 15.0042 KOps/s $\color{#35bf28}+0.28\%$
test_values[td1_return_estimate-False-False] 60.0011ms 58.5491ms 17.0797 Ops/s 18.0587 Ops/s $\textbf{\color{#d91a1a}-5.42\%}$
test_values[vec_td1_return_estimate-False-False] 1.3524ms 1.0663ms 937.8140 Ops/s 930.5250 Ops/s $\color{#35bf28}+0.78\%$
test_values[td_lambda_return_estimate-True-False] 94.6385ms 91.7795ms 10.8957 Ops/s 11.3968 Ops/s $\color{#d91a1a}-4.40\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2794ms 1.0611ms 942.3750 Ops/s 937.0610 Ops/s $\color{#35bf28}+0.57\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 27.0180ms 26.7937ms 37.3222 Ops/s 40.4156 Ops/s $\textbf{\color{#d91a1a}-7.65\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.9418ms 0.7281ms 1.3735 KOps/s 1.3899 KOps/s $\color{#d91a1a}-1.19\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7448ms 0.6761ms 1.4792 KOps/s 1.5260 KOps/s $\color{#d91a1a}-3.07\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5356ms 1.4688ms 680.8062 Ops/s 683.5504 Ops/s $\color{#d91a1a}-0.40\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7559ms 0.7007ms 1.4272 KOps/s 1.4866 KOps/s $\color{#d91a1a}-3.99\%$
test_dqn_speed[False-None] 7.2808ms 1.3447ms 743.6375 Ops/s 742.2919 Ops/s $\color{#35bf28}+0.18\%$
test_dqn_speed[False-backward] 1.9278ms 1.8594ms 537.8136 Ops/s 536.7340 Ops/s $\color{#35bf28}+0.20\%$
test_dqn_speed[True-None] 0.8014ms 0.5734ms 1.7439 KOps/s 1.8054 KOps/s $\color{#d91a1a}-3.40\%$
test_dqn_speed[True-backward] 1.0663ms 1.0144ms 985.8391 Ops/s 954.0237 Ops/s $\color{#35bf28}+3.33\%$
test_dqn_speed[reduce-overhead-None] 0.6690ms 0.5826ms 1.7166 KOps/s 1.7614 KOps/s $\color{#d91a1a}-2.55\%$
test_dqn_speed[reduce-overhead-backward] 1.0726ms 1.0264ms 974.2662 Ops/s 976.1605 Ops/s $\color{#d91a1a}-0.19\%$
test_ddpg_speed[False-None] 3.1916ms 2.7346ms 365.6794 Ops/s 368.1972 Ops/s $\color{#d91a1a}-0.68\%$
test_ddpg_speed[False-backward] 4.2905ms 4.0496ms 246.9369 Ops/s 252.5511 Ops/s $\color{#d91a1a}-2.22\%$
test_ddpg_speed[True-None] 1.3929ms 1.2626ms 791.9916 Ops/s 790.4852 Ops/s $\color{#35bf28}+0.19\%$
test_ddpg_speed[True-backward] 2.3632ms 2.2602ms 442.4402 Ops/s 359.9561 Ops/s $\textbf{\color{#35bf28}+22.92\%}$
test_ddpg_speed[reduce-overhead-None] 1.3984ms 1.2802ms 781.1323 Ops/s 774.3010 Ops/s $\color{#35bf28}+0.88\%$
test_ddpg_speed[reduce-overhead-backward] 2.4499ms 2.2814ms 438.3197 Ops/s 443.1826 Ops/s $\color{#d91a1a}-1.10\%$
test_sac_speed[False-None] 9.0095ms 7.5647ms 132.1922 Ops/s 132.0763 Ops/s $\color{#35bf28}+0.09\%$
test_sac_speed[False-backward] 11.2361ms 10.7755ms 92.8027 Ops/s 93.0893 Ops/s $\color{#d91a1a}-0.31\%$
test_sac_speed[True-None] 2.2559ms 2.0525ms 487.2035 Ops/s 466.4166 Ops/s $\color{#35bf28}+4.46\%$
test_sac_speed[True-backward] 4.0792ms 3.9929ms 250.4446 Ops/s 209.9897 Ops/s $\textbf{\color{#35bf28}+19.27\%}$
test_sac_speed[reduce-overhead-None] 2.2393ms 2.0777ms 481.3007 Ops/s 476.2512 Ops/s $\color{#35bf28}+1.06\%$
test_sac_speed[reduce-overhead-backward] 4.1733ms 4.0293ms 248.1822 Ops/s 247.5540 Ops/s $\color{#35bf28}+0.25\%$
test_redq_speed[False-None] 11.9015ms 10.2486ms 97.5742 Ops/s 98.1495 Ops/s $\color{#d91a1a}-0.59\%$
test_redq_speed[False-backward] 18.9781ms 17.8257ms 56.0988 Ops/s 57.3639 Ops/s $\color{#d91a1a}-2.21\%$
test_redq_speed[True-None] 4.5800ms 3.5563ms 281.1944 Ops/s 281.0855 Ops/s $\color{#35bf28}+0.04\%$
test_redq_speed[True-backward] 9.0130ms 8.6606ms 115.4657 Ops/s 120.2436 Ops/s $\color{#d91a1a}-3.97\%$
test_redq_speed[reduce-overhead-None] 3.8710ms 3.6148ms 276.6399 Ops/s 284.4879 Ops/s $\color{#d91a1a}-2.76\%$
test_redq_speed[reduce-overhead-backward] 9.0447ms 8.6776ms 115.2386 Ops/s 121.3021 Ops/s $\color{#d91a1a}-5.00\%$
test_redq_deprec_speed[False-None] 11.1084ms 10.6354ms 94.0255 Ops/s 95.3216 Ops/s $\color{#d91a1a}-1.36\%$
test_redq_deprec_speed[False-backward] 16.3687ms 15.5547ms 64.2892 Ops/s 65.3380 Ops/s $\color{#d91a1a}-1.61\%$
test_redq_deprec_speed[True-None] 3.5747ms 3.3489ms 298.6041 Ops/s 303.1864 Ops/s $\color{#d91a1a}-1.51\%$
test_redq_deprec_speed[True-backward] 7.3258ms 6.9970ms 142.9190 Ops/s 144.1044 Ops/s $\color{#d91a1a}-0.82\%$
test_redq_deprec_speed[reduce-overhead-None] 3.4293ms 3.2650ms 306.2818 Ops/s 306.1403 Ops/s $\color{#35bf28}+0.05\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.1493ms 6.9271ms 144.3610 Ops/s 144.1907 Ops/s $\color{#35bf28}+0.12\%$
test_td3_speed[False-None] 7.5955ms 7.4788ms 133.7109 Ops/s 133.4609 Ops/s $\color{#35bf28}+0.19\%$
test_td3_speed[False-backward] 11.1663ms 10.4306ms 95.8716 Ops/s 96.4618 Ops/s $\color{#d91a1a}-0.61\%$
test_td3_speed[True-None] 2.1727ms 2.1283ms 469.8641 Ops/s 470.0510 Ops/s $\color{#d91a1a}-0.04\%$
test_td3_speed[True-backward] 4.0668ms 3.9000ms 256.4115 Ops/s 253.7786 Ops/s $\color{#35bf28}+1.04\%$
test_td3_speed[reduce-overhead-None] 2.1556ms 2.1127ms 473.3363 Ops/s 467.9817 Ops/s $\color{#35bf28}+1.14\%$
test_td3_speed[reduce-overhead-backward] 4.6861ms 4.0764ms 245.3117 Ops/s 251.2068 Ops/s $\color{#d91a1a}-2.35\%$
test_cql_speed[False-None] 28.1636ms 25.3008ms 39.5245 Ops/s 40.4194 Ops/s $\color{#d91a1a}-2.21\%$
test_cql_speed[False-backward] 36.8711ms 34.2545ms 29.1932 Ops/s 29.2579 Ops/s $\color{#d91a1a}-0.22\%$
test_cql_speed[True-None] 11.3849ms 10.9542ms 91.2891 Ops/s 93.0159 Ops/s $\color{#d91a1a}-1.86\%$
test_cql_speed[True-backward] 17.5953ms 17.0162ms 58.7674 Ops/s 59.2337 Ops/s $\color{#d91a1a}-0.79\%$
test_cql_speed[reduce-overhead-None] 11.6652ms 11.0068ms 90.8530 Ops/s 90.5290 Ops/s $\color{#35bf28}+0.36\%$
test_cql_speed[reduce-overhead-backward] 18.2602ms 17.3356ms 57.6846 Ops/s 59.4736 Ops/s $\color{#d91a1a}-3.01\%$
test_a2c_speed[False-None] 7.3871ms 5.2637ms 189.9822 Ops/s 184.4984 Ops/s $\color{#35bf28}+2.97\%$
test_a2c_speed[False-backward] 12.2962ms 11.7886ms 84.8277 Ops/s 81.1927 Ops/s $\color{#35bf28}+4.48\%$
test_a2c_speed[True-None] 3.2364ms 3.0916ms 323.4567 Ops/s 316.4420 Ops/s $\color{#35bf28}+2.22\%$
test_a2c_speed[True-backward] 8.6909ms 8.5366ms 117.1430 Ops/s 113.7158 Ops/s $\color{#35bf28}+3.01\%$
test_a2c_speed[reduce-overhead-None] 3.2504ms 3.0983ms 322.7540 Ops/s 321.7106 Ops/s $\color{#35bf28}+0.32\%$
test_a2c_speed[reduce-overhead-backward] 8.9356ms 8.6343ms 115.8178 Ops/s 117.0329 Ops/s $\color{#d91a1a}-1.04\%$
test_ppo_speed[False-None] 5.8979ms 5.6623ms 176.6064 Ops/s 178.7392 Ops/s $\color{#d91a1a}-1.19\%$
test_ppo_speed[False-backward] 12.6081ms 12.3032ms 81.2797 Ops/s 82.3723 Ops/s $\color{#d91a1a}-1.33\%$
test_ppo_speed[True-None] 3.6296ms 3.4651ms 288.5942 Ops/s 285.4024 Ops/s $\color{#35bf28}+1.12\%$
test_ppo_speed[True-backward] 8.6721ms 8.3015ms 120.4600 Ops/s 111.1402 Ops/s $\textbf{\color{#35bf28}+8.39\%}$
test_ppo_speed[reduce-overhead-None] 3.6229ms 3.4652ms 288.5837 Ops/s 287.6504 Ops/s $\color{#35bf28}+0.32\%$
test_ppo_speed[reduce-overhead-backward] 8.5945ms 8.3224ms 120.1571 Ops/s 118.4521 Ops/s $\color{#35bf28}+1.44\%$
test_reinforce_speed[False-None] 6.1766ms 4.4038ms 227.0776 Ops/s 221.1752 Ops/s $\color{#35bf28}+2.67\%$
test_reinforce_speed[False-backward] 7.4937ms 7.2664ms 137.6204 Ops/s 135.4227 Ops/s $\color{#35bf28}+1.62\%$
test_reinforce_speed[True-None] 2.3684ms 2.2172ms 451.0287 Ops/s 443.6302 Ops/s $\color{#35bf28}+1.67\%$
test_reinforce_speed[True-backward] 7.3069ms 7.0743ms 141.3570 Ops/s 133.2421 Ops/s $\textbf{\color{#35bf28}+6.09\%}$
test_reinforce_speed[reduce-overhead-None] 2.5291ms 2.2173ms 450.9934 Ops/s 439.8345 Ops/s $\color{#35bf28}+2.54\%$
test_reinforce_speed[reduce-overhead-backward] 7.4350ms 7.1567ms 139.7291 Ops/s 138.6850 Ops/s $\color{#35bf28}+0.75\%$
test_iql_speed[False-None] 20.4522ms 19.5713ms 51.0953 Ops/s 49.7474 Ops/s $\color{#35bf28}+2.71\%$
test_iql_speed[False-backward] 30.9335ms 30.1479ms 33.1698 Ops/s 33.0824 Ops/s $\color{#35bf28}+0.26\%$
test_iql_speed[True-None] 8.2015ms 7.8831ms 126.8531 Ops/s 122.2158 Ops/s $\color{#35bf28}+3.79\%$
test_iql_speed[True-backward] 17.0615ms 16.5277ms 60.5043 Ops/s 56.2333 Ops/s $\textbf{\color{#35bf28}+7.60\%}$
test_iql_speed[reduce-overhead-None] 8.2901ms 7.9296ms 126.1095 Ops/s 124.2858 Ops/s $\color{#35bf28}+1.47\%$
test_iql_speed[reduce-overhead-backward] 17.0460ms 16.6756ms 59.9678 Ops/s 58.6890 Ops/s $\color{#35bf28}+2.18\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.9543ms 6.7474ms 148.2053 Ops/s 146.6216 Ops/s $\color{#35bf28}+1.08\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.3430s 0.4980ms 2.0082 KOps/s 4.1366 KOps/s $\textbf{\color{#d91a1a}-51.45\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4418ms 0.2190ms 4.5656 KOps/s 4.5384 KOps/s $\color{#35bf28}+0.60\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.0819ms 6.7695ms 147.7206 Ops/s 149.6029 Ops/s $\color{#d91a1a}-1.26\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8040ms 0.2461ms 4.0638 KOps/s 4.1935 KOps/s $\color{#d91a1a}-3.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4337ms 0.2218ms 4.5083 KOps/s 4.6660 KOps/s $\color{#d91a1a}-3.38\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4823ms 1.2575ms 795.2529 Ops/s 757.9897 Ops/s $\color{#35bf28}+4.92\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4122ms 1.2116ms 825.3212 Ops/s 849.4857 Ops/s $\color{#d91a1a}-2.84\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.3231ms 6.9494ms 143.8968 Ops/s 146.4603 Ops/s $\color{#d91a1a}-1.75\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8872ms 0.3929ms 2.5453 KOps/s 2.5951 KOps/s $\color{#d91a1a}-1.92\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6075ms 0.3669ms 2.7258 KOps/s 2.7662 KOps/s $\color{#d91a1a}-1.46\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.0681ms 6.8236ms 146.5497 Ops/s 147.5729 Ops/s $\color{#d91a1a}-0.69\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0690ms 0.2604ms 3.8408 KOps/s 3.9534 KOps/s $\color{#d91a1a}-2.85\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4512ms 0.2521ms 3.9664 KOps/s 3.2250 KOps/s $\textbf{\color{#35bf28}+22.99\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 11.3974ms 6.9136ms 144.6423 Ops/s 150.4670 Ops/s $\color{#d91a1a}-3.87\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0918ms 0.3438ms 2.9090 KOps/s 3.6938 KOps/s $\textbf{\color{#d91a1a}-21.25\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5867ms 0.3225ms 3.1009 KOps/s 3.8686 KOps/s $\textbf{\color{#d91a1a}-19.84\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.2656ms 7.0073ms 142.7087 Ops/s 146.3478 Ops/s $\color{#d91a1a}-2.49\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8823ms 0.3900ms 2.5641 KOps/s 2.0491 KOps/s $\textbf{\color{#35bf28}+25.13\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.5769ms 0.3674ms 2.7220 KOps/s 2.0813 KOps/s $\textbf{\color{#35bf28}+30.78\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.2662ms 5.3445ms 187.1077 Ops/s 34.1792 Ops/s $\textbf{\color{#35bf28}+447.43\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 1.5672ms 1.3687ms 730.6018 Ops/s 638.8672 Ops/s $\textbf{\color{#35bf28}+14.36\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 3.4810ms 1.1259ms 888.1675 Ops/s 812.3635 Ops/s $\textbf{\color{#35bf28}+9.33\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4151s 13.5820ms 73.6267 Ops/s 185.1764 Ops/s $\textbf{\color{#d91a1a}-60.24\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.8115ms 1.9438ms 514.4577 Ops/s 493.2514 Ops/s $\color{#35bf28}+4.30\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.3415ms 1.2975ms 770.7237 Ops/s 882.7042 Ops/s $\textbf{\color{#d91a1a}-12.69\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.6819ms 5.4977ms 181.8953 Ops/s 175.6579 Ops/s $\color{#35bf28}+3.55\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 3.9244ms 1.7070ms 585.8121 Ops/s 464.5913 Ops/s $\textbf{\color{#35bf28}+26.09\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.0726ms 1.3448ms 743.6065 Ops/s 701.5816 Ops/s $\textbf{\color{#35bf28}+5.99\%}$

@thomasbbrunner
Copy link
Contributor

Really nice addition to the docs!

@vmoens vmoens added the documentation Improvements or additions to documentation label Sep 30, 2024
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. documentation Improvements or additions to documentation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants