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

[NOMERG] test 0.7.1 builds #2782

Open
wants to merge 12 commits into
base: main
Choose a base branch
from
Open

[NOMERG] test 0.7.1 builds #2782

wants to merge 12 commits into from

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 12, 2025

  • init
  • [Doc] Fix tutorials ([Doc] Fix tutorials #2768)
  • Update docs.html for BATCHED_PIPE_TIMEOUT
  • init
  • Limit warning filter to torchrl (Limit warning filter to torchrl #2762)
  • [BugFix] Use brackets to get non-tensor data in gym envs
  • [BugFix] Avoid calling reset during env init
  • [BugFix] NonTensor should not convert anything to numpy
  • [BugFix] Fix collector timeouts
  • [BE] Make better logits in cost tests
  • [BugFix] Fix composite setitem
  • [Doc] Solve ref issues in docstrings

Description

Describe your changes in detail.

Motivation and Context

Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
You can use the syntax close #15213 if this solves the issue #15213

  • I have raised an issue to propose this change (required for new features and bug fixes)

Types of changes

What types of changes does your code introduce? Remove all that do not apply:

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds core functionality)
  • Breaking change (fix or feature that would cause existing functionality to change)
  • Documentation (update in the documentation)
  • Example (update in the folder of examples)

Checklist

Go over all the following points, and put an x in all the boxes that apply.
If you are unsure about any of these, don't hesitate to ask. We are here to help!

  • I have read the CONTRIBUTION guide (required)
  • My change requires a change to the documentation.
  • I have updated the tests accordingly (required for a bug fix or a new feature).
  • I have updated the documentation accordingly.

vmoens and others added 11 commits February 5, 2025 21:00
(cherry picked from commit 75f113f)
ghstack-source-id: 3101141eb5b7435c7a4047f5ee84b66c1d74af13
Pull Request resolved: #2769

(cherry picked from commit 84f6b04)
ghstack-source-id: 5ab8281c34aacfd7dbbfc0e285d88bcae0aededf
Pull Request resolved: #2770

(cherry picked from commit 09e93c1)
ghstack-source-id: 7644f6c695490f34d6455703418c59cfa718a9f0
Pull Request resolved: #2771

(cherry picked from commit 3da2750)
ghstack-source-id: cb71d95143beb22db1fe1752e72f70c19f43be79
Pull Request resolved: #2774

(cherry picked from commit f6084b6)
ghstack-source-id: be9ea92b3f3d2592e426eaeaff7b81e50472cf16
Pull Request resolved: #2775

(cherry picked from commit 42ed42c)
ghstack-source-id: f33b49beb4cf8c0c8b156559b1abbee8ac77db20
Pull Request resolved: #2778

(cherry picked from commit c2a149d)
Copy link

pytorch-bot bot commented Feb 12, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2782

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

❌ 2 New Failures, 2 Unrelated Failures

As of commit 36d260c with merge base 4c06ce2 (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.

@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 Feb 12, 2025
@vmoens vmoens force-pushed the release/0.7.1 branch 7 times, most recently from 6fea6c4 to efd2351 Compare February 12, 2025 21:11
ghstack-source-id: 09823fa85a94115291e7434478776fb0834f9b39
Pull Request resolved: #2776

(cherry picked from commit f5445a4)
Copy link

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

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

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.6044s 0.5108s 1.9578 Ops/s 2.0258 Ops/s $\color{#d91a1a}-3.35\%$
test_transformed 1.0971s 0.9961s 1.0039 Ops/s 1.0272 Ops/s $\color{#d91a1a}-2.27\%$
test_serial 1.6114s 1.5207s 0.6576 Ops/s 0.6734 Ops/s $\color{#d91a1a}-2.35\%$
test_parallel 1.4138s 1.3030s 0.7674 Ops/s 0.7737 Ops/s $\color{#d91a1a}-0.80\%$
test_step_mdp_speed[True-True-True-True-True] 0.2551ms 30.1112μs 33.2102 KOps/s 32.3386 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[True-True-True-True-False] 63.2380μs 17.7988μs 56.1837 KOps/s 55.4888 KOps/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-True-True-False-True] 80.5600μs 16.9597μs 58.9633 KOps/s 58.2179 KOps/s $\color{#35bf28}+1.28\%$
test_step_mdp_speed[True-True-True-False-False] 39.0430μs 9.9150μs 100.8577 KOps/s 98.9680 KOps/s $\color{#35bf28}+1.91\%$
test_step_mdp_speed[True-True-False-True-True] 97.8430μs 32.1712μs 31.0837 KOps/s 31.0226 KOps/s $\color{#35bf28}+0.20\%$
test_step_mdp_speed[True-True-False-True-False] 82.1330μs 19.6181μs 50.9733 KOps/s 50.3269 KOps/s $\color{#35bf28}+1.28\%$
test_step_mdp_speed[True-True-False-False-True] 73.2570μs 19.1862μs 52.1207 KOps/s 52.6959 KOps/s $\color{#d91a1a}-1.09\%$
test_step_mdp_speed[True-True-False-False-False] 67.1960μs 11.7593μs 85.0387 KOps/s 83.3474 KOps/s $\color{#35bf28}+2.03\%$
test_step_mdp_speed[True-False-True-True-True] 79.8590μs 34.3515μs 29.1108 KOps/s 29.1308 KOps/s $\color{#d91a1a}-0.07\%$
test_step_mdp_speed[True-False-True-True-False] 74.4090μs 21.3596μs 46.8174 KOps/s 45.8796 KOps/s $\color{#35bf28}+2.04\%$
test_step_mdp_speed[True-False-True-False-True] 87.0730μs 18.7082μs 53.4525 KOps/s 52.4412 KOps/s $\color{#35bf28}+1.93\%$
test_step_mdp_speed[True-False-True-False-False] 38.9930μs 11.7797μs 84.8918 KOps/s 83.6324 KOps/s $\color{#35bf28}+1.51\%$
test_step_mdp_speed[True-False-False-True-True] 97.0110μs 35.6282μs 28.0676 KOps/s 28.0182 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[True-False-False-True-False] 55.2930μs 23.1985μs 43.1063 KOps/s 42.8625 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[True-False-False-False-True] 76.9340μs 20.4177μs 48.9771 KOps/s 48.1462 KOps/s $\color{#35bf28}+1.73\%$
test_step_mdp_speed[True-False-False-False-False] 76.1930μs 13.5490μs 73.8063 KOps/s 72.8377 KOps/s $\color{#35bf28}+1.33\%$
test_step_mdp_speed[False-True-True-True-True] 78.3770μs 34.1686μs 29.2666 KOps/s 29.2653 KOps/s $+0.00\%$
test_step_mdp_speed[False-True-True-True-False] 0.5241ms 21.6899μs 46.1044 KOps/s 46.0982 KOps/s $\color{#35bf28}+0.01\%$
test_step_mdp_speed[False-True-True-False-True] 56.7860μs 21.5880μs 46.3220 KOps/s 46.0522 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[False-True-True-False-False] 67.8070μs 13.1409μs 76.0983 KOps/s 75.7460 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[False-True-False-True-True] 0.1063ms 35.6367μs 28.0610 KOps/s 28.0672 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[False-True-False-True-False] 52.6790μs 23.4208μs 42.6972 KOps/s 42.3541 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[False-True-False-False-True] 2.8844ms 23.2842μs 42.9476 KOps/s 42.7226 KOps/s $\color{#35bf28}+0.53\%$
test_step_mdp_speed[False-True-False-False-False] 47.8600μs 14.9813μs 66.7500 KOps/s 66.2184 KOps/s $\color{#35bf28}+0.80\%$
test_step_mdp_speed[False-False-True-True-True] 0.1048ms 37.1142μs 26.9439 KOps/s 26.4505 KOps/s $\color{#35bf28}+1.87\%$
test_step_mdp_speed[False-False-True-True-False] 80.1500μs 25.1126μs 39.8206 KOps/s 39.1466 KOps/s $\color{#35bf28}+1.72\%$
test_step_mdp_speed[False-False-True-False-True] 49.7830μs 23.6775μs 42.2342 KOps/s 42.7009 KOps/s $\color{#d91a1a}-1.09\%$
test_step_mdp_speed[False-False-True-False-False] 98.8450μs 15.0123μs 66.6121 KOps/s 65.9037 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-False-False-True-True] 74.8910μs 39.0869μs 25.5840 KOps/s 25.5941 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[False-False-False-True-False] 80.7810μs 26.6685μs 37.4974 KOps/s 37.2332 KOps/s $\color{#35bf28}+0.71\%$
test_step_mdp_speed[False-False-False-False-True] 89.4470μs 24.6877μs 40.5060 KOps/s 38.9778 KOps/s $\color{#35bf28}+3.92\%$
test_step_mdp_speed[False-False-False-False-False] 0.1838ms 16.6890μs 59.9198 KOps/s 58.8169 KOps/s $\color{#35bf28}+1.88\%$
test_values[generalized_advantage_estimate-True-True] 13.3438ms 10.1840ms 98.1931 Ops/s 104.1091 Ops/s $\textbf{\color{#d91a1a}-5.68\%}$
test_values[vec_generalized_advantage_estimate-True-True] 29.4468ms 26.2280ms 38.1272 Ops/s 37.2746 Ops/s $\color{#35bf28}+2.29\%$
test_values[td0_return_estimate-False-False] 0.2506ms 0.1793ms 5.5782 KOps/s 5.4759 KOps/s $\color{#35bf28}+1.87\%$
test_values[td1_return_estimate-False-False] 24.9558ms 24.4721ms 40.8629 Ops/s 42.9253 Ops/s $\color{#d91a1a}-4.80\%$
test_values[vec_td1_return_estimate-False-False] 35.5978ms 26.7135ms 37.4342 Ops/s 37.8360 Ops/s $\color{#d91a1a}-1.06\%$
test_values[td_lambda_return_estimate-True-False] 38.9264ms 35.8646ms 27.8827 Ops/s 29.8619 Ops/s $\textbf{\color{#d91a1a}-6.63\%}$
test_values[vec_td_lambda_return_estimate-True-False] 27.4049ms 26.5477ms 37.6681 Ops/s 37.9868 Ops/s $\color{#d91a1a}-0.84\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.7980ms 8.7327ms 114.5119 Ops/s 120.7847 Ops/s $\textbf{\color{#d91a1a}-5.19\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4546ms 1.9166ms 521.7646 Ops/s 496.7692 Ops/s $\textbf{\color{#35bf28}+5.03\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4989ms 0.3694ms 2.7073 KOps/s 2.7634 KOps/s $\color{#d91a1a}-2.03\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 48.7957ms 45.4426ms 22.0058 Ops/s 22.7562 Ops/s $\color{#d91a1a}-3.30\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.2662ms 3.4776ms 287.5566 Ops/s 291.9923 Ops/s $\color{#d91a1a}-1.52\%$
test_dqn_speed[False-None] 6.3585ms 1.4775ms 676.8153 Ops/s 700.2643 Ops/s $\color{#d91a1a}-3.35\%$
test_dqn_speed[False-backward] 2.0320ms 1.9547ms 511.5853 Ops/s 524.9528 Ops/s $\color{#d91a1a}-2.55\%$
test_dqn_speed[True-None] 0.6431ms 0.4888ms 2.0457 KOps/s 2.0789 KOps/s $\color{#d91a1a}-1.60\%$
test_dqn_speed[True-backward] 1.0070ms 0.9220ms 1.0846 KOps/s 801.1685 Ops/s $\textbf{\color{#35bf28}+35.37\%}$
test_dqn_speed[reduce-overhead-None] 0.7505ms 0.4900ms 2.0408 KOps/s 2.0721 KOps/s $\color{#d91a1a}-1.51\%$
test_dqn_speed[reduce-overhead-backward] 0.9611ms 0.9128ms 1.0955 KOps/s 1.0976 KOps/s $\color{#d91a1a}-0.19\%$
test_ddpg_speed[False-None] 3.3448ms 2.9142ms 343.1441 Ops/s 340.7970 Ops/s $\color{#35bf28}+0.69\%$
test_ddpg_speed[False-backward] 5.1661ms 4.0885ms 244.5872 Ops/s 250.2243 Ops/s $\color{#d91a1a}-2.25\%$
test_ddpg_speed[True-None] 1.9209ms 1.2382ms 807.6040 Ops/s 804.4264 Ops/s $\color{#35bf28}+0.40\%$
test_ddpg_speed[True-backward] 2.1621ms 2.1103ms 473.8771 Ops/s 473.8900 Ops/s $-0.00\%$
test_ddpg_speed[reduce-overhead-None] 1.6333ms 1.2391ms 807.0086 Ops/s 805.2878 Ops/s $\color{#35bf28}+0.21\%$
test_ddpg_speed[reduce-overhead-backward] 2.2140ms 2.1140ms 473.0320 Ops/s 472.0207 Ops/s $\color{#35bf28}+0.21\%$
test_sac_speed[False-None] 8.7444ms 8.1020ms 123.4264 Ops/s 123.6294 Ops/s $\color{#d91a1a}-0.16\%$
test_sac_speed[False-backward] 13.2766ms 11.1079ms 90.0261 Ops/s 92.0834 Ops/s $\color{#d91a1a}-2.23\%$
test_sac_speed[True-None] 2.3344ms 2.0958ms 477.1422 Ops/s 470.7088 Ops/s $\color{#35bf28}+1.37\%$
test_sac_speed[True-backward] 3.8469ms 3.7644ms 265.6478 Ops/s 267.1427 Ops/s $\color{#d91a1a}-0.56\%$
test_sac_speed[reduce-overhead-None] 2.6127ms 2.0946ms 477.4101 Ops/s 473.2796 Ops/s $\color{#35bf28}+0.87\%$
test_sac_speed[reduce-overhead-backward] 3.8448ms 3.7637ms 265.6946 Ops/s 258.5503 Ops/s $\color{#35bf28}+2.76\%$
test_redq_speed[False-None] 19.7471ms 13.8806ms 72.0432 Ops/s 75.9876 Ops/s $\textbf{\color{#d91a1a}-5.19\%}$
test_redq_speed[False-backward] 33.6436ms 23.6733ms 42.2417 Ops/s 44.2318 Ops/s $\color{#d91a1a}-4.50\%$
test_redq_speed[True-None] 6.4245ms 5.2972ms 188.7804 Ops/s 192.6589 Ops/s $\color{#d91a1a}-2.01\%$
test_redq_speed[True-backward] 15.1867ms 13.7057ms 72.9624 Ops/s 75.1288 Ops/s $\color{#d91a1a}-2.88\%$
test_redq_speed[reduce-overhead-None] 6.2515ms 5.4219ms 184.4368 Ops/s 190.5695 Ops/s $\color{#d91a1a}-3.22\%$
test_redq_speed[reduce-overhead-backward] 14.6868ms 13.1464ms 76.0665 Ops/s 75.7421 Ops/s $\color{#35bf28}+0.43\%$
test_redq_deprec_speed[False-None] 16.4057ms 13.2941ms 75.2212 Ops/s 75.4495 Ops/s $\color{#d91a1a}-0.30\%$
test_redq_deprec_speed[False-backward] 20.7267ms 18.7462ms 53.3443 Ops/s 52.5780 Ops/s $\color{#35bf28}+1.46\%$
test_redq_deprec_speed[True-None] 4.4033ms 3.9123ms 255.6018 Ops/s 232.0597 Ops/s $\textbf{\color{#35bf28}+10.14\%}$
test_redq_deprec_speed[True-backward] 9.5346ms 8.6801ms 115.2067 Ops/s 104.8457 Ops/s $\textbf{\color{#35bf28}+9.88\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.7536ms 3.9376ms 253.9627 Ops/s 237.2853 Ops/s $\textbf{\color{#35bf28}+7.03\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.8346ms 8.7472ms 114.3225 Ops/s 112.7774 Ops/s $\color{#35bf28}+1.37\%$
test_td3_speed[False-None] 8.4684ms 8.1994ms 121.9601 Ops/s 121.0590 Ops/s $\color{#35bf28}+0.74\%$
test_td3_speed[False-backward] 12.2538ms 10.6953ms 93.4991 Ops/s 93.5896 Ops/s $\color{#d91a1a}-0.10\%$
test_td3_speed[True-None] 2.0728ms 1.8192ms 549.6972 Ops/s 543.5741 Ops/s $\color{#35bf28}+1.13\%$
test_td3_speed[True-backward] 3.6469ms 3.5105ms 284.8618 Ops/s 288.8950 Ops/s $\color{#d91a1a}-1.40\%$
test_td3_speed[reduce-overhead-None] 2.0570ms 1.8349ms 544.9903 Ops/s 541.4803 Ops/s $\color{#35bf28}+0.65\%$
test_td3_speed[reduce-overhead-backward] 3.6359ms 3.4451ms 290.2664 Ops/s 288.3754 Ops/s $\color{#35bf28}+0.66\%$
test_cql_speed[False-None] 38.0312ms 36.9247ms 27.0822 Ops/s 26.8173 Ops/s $\color{#35bf28}+0.99\%$
test_cql_speed[False-backward] 50.8468ms 48.0401ms 20.8160 Ops/s 20.9849 Ops/s $\color{#d91a1a}-0.81\%$
test_cql_speed[True-None] 18.5833ms 16.8439ms 59.3688 Ops/s 59.3490 Ops/s $\color{#35bf28}+0.03\%$
test_cql_speed[True-backward] 26.3211ms 24.0299ms 41.6149 Ops/s 42.4032 Ops/s $\color{#d91a1a}-1.86\%$
test_cql_speed[reduce-overhead-None] 17.8211ms 16.8248ms 59.4361 Ops/s 61.0914 Ops/s $\color{#d91a1a}-2.71\%$
test_cql_speed[reduce-overhead-backward] 26.1031ms 23.9669ms 41.7242 Ops/s 42.5697 Ops/s $\color{#d91a1a}-1.99\%$
test_a2c_speed[False-None] 8.7190ms 7.5401ms 132.6246 Ops/s 137.8560 Ops/s $\color{#d91a1a}-3.79\%$
test_a2c_speed[False-backward] 15.5756ms 15.1525ms 65.9955 Ops/s 68.2924 Ops/s $\color{#d91a1a}-3.36\%$
test_a2c_speed[True-None] 4.1675ms 3.8012ms 263.0740 Ops/s 267.3360 Ops/s $\color{#d91a1a}-1.59\%$
test_a2c_speed[True-backward] 12.2331ms 10.8577ms 92.1008 Ops/s 96.6715 Ops/s $\color{#d91a1a}-4.73\%$
test_a2c_speed[reduce-overhead-None] 4.2745ms 3.8121ms 262.3245 Ops/s 267.2509 Ops/s $\color{#d91a1a}-1.84\%$
test_a2c_speed[reduce-overhead-backward] 11.2846ms 10.6579ms 93.8274 Ops/s 90.5498 Ops/s $\color{#35bf28}+3.62\%$
test_ppo_speed[False-None] 8.7595ms 7.8351ms 127.6315 Ops/s 129.2218 Ops/s $\color{#d91a1a}-1.23\%$
test_ppo_speed[False-backward] 17.1606ms 15.7236ms 63.5987 Ops/s 66.1855 Ops/s $\color{#d91a1a}-3.91\%$
test_ppo_speed[True-None] 5.2642ms 4.2109ms 237.4773 Ops/s 243.3175 Ops/s $\color{#d91a1a}-2.40\%$
test_ppo_speed[True-backward] 11.8875ms 10.6074ms 94.2736 Ops/s 96.6798 Ops/s $\color{#d91a1a}-2.49\%$
test_ppo_speed[reduce-overhead-None] 4.4247ms 4.1370ms 241.7232 Ops/s 240.4353 Ops/s $\color{#35bf28}+0.54\%$
test_ppo_speed[reduce-overhead-backward] 10.5643ms 10.1739ms 98.2904 Ops/s 96.9513 Ops/s $\color{#35bf28}+1.38\%$
test_reinforce_speed[False-None] 7.5858ms 6.6333ms 150.7542 Ops/s 151.1869 Ops/s $\color{#d91a1a}-0.29\%$
test_reinforce_speed[False-backward] 10.8601ms 9.9770ms 100.2310 Ops/s 100.6485 Ops/s $\color{#d91a1a}-0.41\%$
test_reinforce_speed[True-None] 4.0750ms 3.1099ms 321.5541 Ops/s 323.5620 Ops/s $\color{#d91a1a}-0.62\%$
test_reinforce_speed[True-backward] 10.2944ms 9.1818ms 108.9113 Ops/s 106.8657 Ops/s $\color{#35bf28}+1.91\%$
test_reinforce_speed[reduce-overhead-None] 3.5877ms 3.1153ms 320.9987 Ops/s 316.2163 Ops/s $\color{#35bf28}+1.51\%$
test_reinforce_speed[reduce-overhead-backward] 10.9017ms 9.2053ms 108.6328 Ops/s 109.8309 Ops/s $\color{#d91a1a}-1.09\%$
test_iql_speed[False-None] 34.5575ms 32.9236ms 30.3734 Ops/s 30.6994 Ops/s $\color{#d91a1a}-1.06\%$
test_iql_speed[False-backward] 49.1663ms 46.5725ms 21.4719 Ops/s 21.6872 Ops/s $\color{#d91a1a}-0.99\%$
test_iql_speed[True-None] 13.4056ms 12.0186ms 83.2041 Ops/s 85.2536 Ops/s $\color{#d91a1a}-2.40\%$
test_iql_speed[True-backward] 25.3158ms 23.0613ms 43.3627 Ops/s 44.1329 Ops/s $\color{#d91a1a}-1.75\%$
test_iql_speed[reduce-overhead-None] 12.4827ms 11.5914ms 86.2710 Ops/s 86.8288 Ops/s $\color{#d91a1a}-0.64\%$
test_iql_speed[reduce-overhead-backward] 24.4455ms 22.9426ms 43.5870 Ops/s 41.9819 Ops/s $\color{#35bf28}+3.82\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.4384ms 5.0388ms 198.4601 Ops/s 190.7464 Ops/s $\color{#35bf28}+4.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8756ms 0.5297ms 1.8877 KOps/s 1.7792 KOps/s $\textbf{\color{#35bf28}+6.09\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9850ms 0.4999ms 2.0004 KOps/s 1.9330 KOps/s $\color{#35bf28}+3.49\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.2757ms 4.8168ms 207.6077 Ops/s 202.4964 Ops/s $\color{#35bf28}+2.52\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8337ms 0.5160ms 1.9379 KOps/s 1.9207 KOps/s $\color{#35bf28}+0.90\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7894ms 0.4879ms 2.0494 KOps/s 1.9974 KOps/s $\color{#35bf28}+2.60\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.5083ms 1.6685ms 599.3378 Ops/s 606.9976 Ops/s $\color{#d91a1a}-1.26\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.2925ms 1.5847ms 631.0203 Ops/s 637.4660 Ops/s $\color{#d91a1a}-1.01\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.9602ms 4.9027ms 203.9710 Ops/s 205.5577 Ops/s $\color{#d91a1a}-0.77\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2253ms 0.6631ms 1.5081 KOps/s 1.5140 KOps/s $\color{#d91a1a}-0.39\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0125ms 0.6386ms 1.5659 KOps/s 1.5761 KOps/s $\color{#d91a1a}-0.65\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 4.9843ms 4.7331ms 211.2770 Ops/s 215.4672 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6165s 1.5447ms 647.3746 Ops/s 1.8875 KOps/s $\textbf{\color{#d91a1a}-65.70\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8634ms 0.5098ms 1.9616 KOps/s 2.0074 KOps/s $\color{#d91a1a}-2.28\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.5148ms 4.8267ms 207.1806 Ops/s 214.7174 Ops/s $\color{#d91a1a}-3.51\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0825ms 0.5124ms 1.9515 KOps/s 407.5629 Ops/s $\textbf{\color{#35bf28}+378.83\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9133ms 0.5011ms 1.9958 KOps/s 2.0244 KOps/s $\color{#d91a1a}-1.41\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.5595ms 4.9521ms 201.9331 Ops/s 201.8033 Ops/s $\color{#35bf28}+0.06\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1459ms 0.6632ms 1.5078 KOps/s 1.5103 KOps/s $\color{#d91a1a}-0.17\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8527ms 0.6355ms 1.5735 KOps/s 1.5863 KOps/s $\color{#d91a1a}-0.80\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.6223ms 4.3679ms 228.9444 Ops/s 228.9317 Ops/s $+0.01\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.1745ms 2.4106ms 414.8367 Ops/s 449.1030 Ops/s $\textbf{\color{#d91a1a}-7.63\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.5107ms 1.3689ms 730.5209 Ops/s 735.6920 Ops/s $\color{#d91a1a}-0.70\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4788s 13.9384ms 71.7441 Ops/s 230.7445 Ops/s $\textbf{\color{#d91a1a}-68.91\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.5123ms 2.2877ms 437.1218 Ops/s 424.7790 Ops/s $\color{#35bf28}+2.91\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.4881ms 1.3689ms 730.5356 Ops/s 735.9142 Ops/s $\color{#d91a1a}-0.73\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.0006ms 4.4323ms 225.6181 Ops/s 218.4649 Ops/s $\color{#35bf28}+3.27\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 7.4317ms 2.4627ms 406.0518 Ops/s 388.7242 Ops/s $\color{#35bf28}+4.46\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 5.5223ms 1.4753ms 677.8089 Ops/s 692.1781 Ops/s $\color{#d91a1a}-2.08\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.8118ms 12.2559ms 81.5933 Ops/s 84.6148 Ops/s $\color{#d91a1a}-3.57\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.9221ms 14.8965ms 67.1299 Ops/s 69.4867 Ops/s $\color{#d91a1a}-3.39\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 23.2693ms 20.8545ms 47.9512 Ops/s 47.8371 Ops/s $\color{#35bf28}+0.24\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.1016ms 14.9092ms 67.0729 Ops/s 68.6447 Ops/s $\color{#d91a1a}-2.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.1529ms 20.7743ms 48.1364 Ops/s 47.8339 Ops/s $\color{#35bf28}+0.63\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.1837ms 16.3901ms 61.0124 Ops/s 63.3233 Ops/s $\color{#d91a1a}-3.65\%$

Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}23$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8808s 0.7961s 1.2561 Ops/s 1.2587 Ops/s $\color{#d91a1a}-0.21\%$
test_transformed 1.4626s 1.3783s 0.7255 Ops/s 0.7287 Ops/s $\color{#d91a1a}-0.43\%$
test_serial 2.3760s 2.2920s 0.4363 Ops/s 0.4359 Ops/s $\color{#35bf28}+0.10\%$
test_parallel 1.9057s 1.8463s 0.5416 Ops/s 0.5358 Ops/s $\color{#35bf28}+1.08\%$
test_step_mdp_speed[True-True-True-True-True] 0.2428ms 40.5675μs 24.6503 KOps/s 24.3620 KOps/s $\color{#35bf28}+1.18\%$
test_step_mdp_speed[True-True-True-True-False] 61.7100μs 23.5984μs 42.3758 KOps/s 41.9351 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-True-True-False-True] 61.6210μs 22.1762μs 45.0933 KOps/s 44.7697 KOps/s $\color{#35bf28}+0.72\%$
test_step_mdp_speed[True-True-True-False-False] 37.8600μs 12.9313μs 77.3319 KOps/s 78.3284 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-True-False-True-True] 0.1244ms 42.8452μs 23.3398 KOps/s 23.3244 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[True-True-False-True-False] 70.1110μs 25.3153μs 39.5018 KOps/s 39.1678 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[True-True-False-False-True] 61.5910μs 24.5939μs 40.6605 KOps/s 40.0326 KOps/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[True-True-False-False-False] 0.4630ms 15.4214μs 64.8449 KOps/s 64.8945 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[True-False-True-True-True] 90.7510μs 45.1514μs 22.1477 KOps/s 22.0822 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[True-False-True-True-False] 0.4259ms 28.0213μs 35.6871 KOps/s 35.3057 KOps/s $\color{#35bf28}+1.08\%$
test_step_mdp_speed[True-False-True-False-True] 0.4383ms 24.5066μs 40.8053 KOps/s 39.6292 KOps/s $\color{#35bf28}+2.97\%$
test_step_mdp_speed[True-False-True-False-False] 0.4166ms 15.3902μs 64.9765 KOps/s 64.4716 KOps/s $\color{#35bf28}+0.78\%$
test_step_mdp_speed[True-False-False-True-True] 83.2010μs 46.8467μs 21.3462 KOps/s 20.9446 KOps/s $\color{#35bf28}+1.92\%$
test_step_mdp_speed[True-False-False-True-False] 0.4466ms 30.0288μs 33.3014 KOps/s 32.9547 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-False-False-False-True] 63.2210μs 26.8477μs 37.2472 KOps/s 37.1778 KOps/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[True-False-False-False-False] 0.4225ms 17.4643μs 57.2597 KOps/s 57.4534 KOps/s $\color{#d91a1a}-0.34\%$
test_step_mdp_speed[False-True-True-True-True] 0.4368ms 45.2140μs 22.1171 KOps/s 22.3491 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[False-True-True-True-False] 61.9300μs 28.1148μs 35.5685 KOps/s 35.5121 KOps/s $\color{#35bf28}+0.16\%$
test_step_mdp_speed[False-True-True-False-True] 0.4373ms 28.0122μs 35.6987 KOps/s 35.7273 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[False-True-True-False-False] 0.4112ms 17.0755μs 58.5634 KOps/s 59.1061 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[False-True-False-True-True] 0.4423ms 47.8699μs 20.8900 KOps/s 21.1798 KOps/s $\color{#d91a1a}-1.37\%$
test_step_mdp_speed[False-True-False-True-False] 65.1110μs 30.1420μs 33.1763 KOps/s 33.1489 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[False-True-False-False-True] 3.2534ms 30.9812μs 32.2776 KOps/s 31.9952 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[False-True-False-False-False] 0.4300ms 19.0107μs 52.6020 KOps/s 51.5034 KOps/s $\color{#35bf28}+2.13\%$
test_step_mdp_speed[False-False-True-True-True] 0.4415ms 49.6985μs 20.1213 KOps/s 20.0593 KOps/s $\color{#35bf28}+0.31\%$
test_step_mdp_speed[False-False-True-True-False] 0.4197ms 33.2608μs 30.0654 KOps/s 30.6355 KOps/s $\color{#d91a1a}-1.86\%$
test_step_mdp_speed[False-False-True-False-True] 68.9610μs 30.8349μs 32.4308 KOps/s 32.1642 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-False-True-False-False] 0.4091ms 19.1127μs 52.3213 KOps/s 51.8293 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[False-False-False-True-True] 0.4634ms 51.0111μs 19.6036 KOps/s 19.4438 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-False-False-True-False] 0.4204ms 35.1520μs 28.4479 KOps/s 28.7798 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[False-False-False-False-True] 69.9510μs 32.2046μs 31.0515 KOps/s 30.8466 KOps/s $\color{#35bf28}+0.66\%$
test_step_mdp_speed[False-False-False-False-False] 0.4273ms 21.1374μs 47.3094 KOps/s 46.6587 KOps/s $\color{#35bf28}+1.39\%$
test_values[generalized_advantage_estimate-True-True] 25.0946ms 24.6445ms 40.5771 Ops/s 40.0396 Ops/s $\color{#35bf28}+1.34\%$
test_values[vec_generalized_advantage_estimate-True-True] 98.1242ms 2.8710ms 348.3104 Ops/s 327.4039 Ops/s $\textbf{\color{#35bf28}+6.39\%}$
test_values[td0_return_estimate-False-False] 0.1062ms 80.8086μs 12.3749 KOps/s 12.1957 KOps/s $\color{#35bf28}+1.47\%$
test_values[td1_return_estimate-False-False] 63.8652ms 55.5212ms 18.0111 Ops/s 17.8915 Ops/s $\color{#35bf28}+0.67\%$
test_values[vec_td1_return_estimate-False-False] 1.3574ms 1.0885ms 918.7345 Ops/s 911.4247 Ops/s $\color{#35bf28}+0.80\%$
test_values[td_lambda_return_estimate-True-False] 88.8419ms 88.1949ms 11.3385 Ops/s 11.2742 Ops/s $\color{#35bf28}+0.57\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3067ms 1.0883ms 918.8478 Ops/s 917.5439 Ops/s $\color{#35bf28}+0.14\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 24.8861ms 24.7485ms 40.4065 Ops/s 40.1498 Ops/s $\color{#35bf28}+0.64\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0270ms 0.7556ms 1.3234 KOps/s 1.3064 KOps/s $\color{#35bf28}+1.30\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7869ms 0.6749ms 1.4816 KOps/s 1.4693 KOps/s $\color{#35bf28}+0.84\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5362ms 1.4859ms 673.0145 Ops/s 668.0377 Ops/s $\color{#35bf28}+0.74\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7571ms 0.6909ms 1.4475 KOps/s 1.4079 KOps/s $\color{#35bf28}+2.81\%$
test_dqn_speed[False-None] 7.1321ms 1.5568ms 642.3228 Ops/s 650.1124 Ops/s $\color{#d91a1a}-1.20\%$
test_dqn_speed[False-backward] 2.2517ms 2.1577ms 463.4535 Ops/s 462.8384 Ops/s $\color{#35bf28}+0.13\%$
test_dqn_speed[True-None] 0.7014ms 0.5557ms 1.7996 KOps/s 1.7762 KOps/s $\color{#35bf28}+1.32\%$
test_dqn_speed[True-backward] 1.2981ms 1.2311ms 812.2619 Ops/s 804.6728 Ops/s $\color{#35bf28}+0.94\%$
test_dqn_speed[reduce-overhead-None] 0.9839ms 0.5715ms 1.7497 KOps/s 1.7245 KOps/s $\color{#35bf28}+1.46\%$
test_dqn_speed[reduce-overhead-backward] 1.1126ms 1.0673ms 936.9499 Ops/s 919.7110 Ops/s $\color{#35bf28}+1.87\%$
test_ddpg_speed[False-None] 3.3021ms 2.9162ms 342.9178 Ops/s 342.4910 Ops/s $\color{#35bf28}+0.12\%$
test_ddpg_speed[False-backward] 4.7017ms 4.2836ms 233.4464 Ops/s 232.1372 Ops/s $\color{#35bf28}+0.56\%$
test_ddpg_speed[True-None] 1.7492ms 1.3418ms 745.2868 Ops/s 738.8121 Ops/s $\color{#35bf28}+0.88\%$
test_ddpg_speed[True-backward] 3.0087ms 2.5817ms 387.3460 Ops/s 379.5445 Ops/s $\color{#35bf28}+2.06\%$
test_ddpg_speed[reduce-overhead-None] 1.7914ms 1.3758ms 726.8325 Ops/s 734.1305 Ops/s $\color{#d91a1a}-0.99\%$
test_ddpg_speed[reduce-overhead-backward] 2.2413ms 2.1082ms 474.3441 Ops/s 483.6979 Ops/s $\color{#d91a1a}-1.93\%$
test_sac_speed[False-None] 8.5945ms 8.1796ms 122.2553 Ops/s 122.5653 Ops/s $\color{#d91a1a}-0.25\%$
test_sac_speed[False-backward] 12.0948ms 11.3540ms 88.0750 Ops/s 87.8856 Ops/s $\color{#35bf28}+0.22\%$
test_sac_speed[True-None] 2.0350ms 1.8902ms 529.0370 Ops/s 535.5156 Ops/s $\color{#d91a1a}-1.21\%$
test_sac_speed[True-backward] 3.9854ms 3.7548ms 266.3272 Ops/s 262.6016 Ops/s $\color{#35bf28}+1.42\%$
test_sac_speed[reduce-overhead-None] 18.3008ms 10.9195ms 91.5792 Ops/s 91.5796 Ops/s $-0.00\%$
test_sac_speed[reduce-overhead-backward] 1.8649ms 1.7955ms 556.9464 Ops/s 546.9795 Ops/s $\color{#35bf28}+1.82\%$
test_redq_speed[False-None] 7.9549ms 7.4643ms 133.9703 Ops/s 130.4541 Ops/s $\color{#35bf28}+2.70\%$
test_redq_speed[False-backward] 12.2283ms 11.7378ms 85.1947 Ops/s 83.9568 Ops/s $\color{#35bf28}+1.47\%$
test_redq_speed[True-None] 2.3825ms 2.3107ms 432.7623 Ops/s 418.9740 Ops/s $\color{#35bf28}+3.29\%$
test_redq_speed[True-backward] 4.4773ms 4.0249ms 248.4505 Ops/s 231.9015 Ops/s $\textbf{\color{#35bf28}+7.14\%}$
test_redq_speed[reduce-overhead-None] 2.3989ms 2.3416ms 427.0601 Ops/s 415.9493 Ops/s $\color{#35bf28}+2.67\%$
test_redq_speed[reduce-overhead-backward] 4.5332ms 4.0488ms 246.9870 Ops/s 231.6398 Ops/s $\textbf{\color{#35bf28}+6.63\%}$
test_redq_deprec_speed[False-None] 9.3949ms 9.0803ms 110.1287 Ops/s 109.3083 Ops/s $\color{#35bf28}+0.75\%$
test_redq_deprec_speed[False-backward] 12.5071ms 12.0346ms 83.0941 Ops/s 80.2222 Ops/s $\color{#35bf28}+3.58\%$
test_redq_deprec_speed[True-None] 2.7406ms 2.6278ms 380.5479 Ops/s 370.8209 Ops/s $\color{#35bf28}+2.62\%$
test_redq_deprec_speed[True-backward] 4.6850ms 4.2966ms 232.7402 Ops/s 217.0342 Ops/s $\textbf{\color{#35bf28}+7.24\%}$
test_redq_deprec_speed[reduce-overhead-None] 2.7428ms 2.6420ms 378.5011 Ops/s 369.5417 Ops/s $\color{#35bf28}+2.42\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.7759ms 4.2938ms 232.8961 Ops/s 216.9962 Ops/s $\textbf{\color{#35bf28}+7.33\%}$
test_td3_speed[False-None] 8.0383ms 7.9941ms 125.0929 Ops/s 122.7346 Ops/s $\color{#35bf28}+1.92\%$
test_td3_speed[False-backward] 10.8351ms 10.3367ms 96.7428 Ops/s 93.5987 Ops/s $\color{#35bf28}+3.36\%$
test_td3_speed[True-None] 1.6708ms 1.6467ms 607.2754 Ops/s 594.4611 Ops/s $\color{#35bf28}+2.16\%$
test_td3_speed[True-backward] 3.2196ms 3.1809ms 314.3747 Ops/s 292.3378 Ops/s $\textbf{\color{#35bf28}+7.54\%}$
test_td3_speed[reduce-overhead-None] 69.5422ms 26.9885ms 37.0528 Ops/s 37.2361 Ops/s $\color{#d91a1a}-0.49\%$
test_td3_speed[reduce-overhead-backward] 1.3931ms 1.3523ms 739.4766 Ops/s 654.2294 Ops/s $\textbf{\color{#35bf28}+13.03\%}$
test_cql_speed[False-None] 17.3172ms 16.8463ms 59.3602 Ops/s 58.2652 Ops/s $\color{#35bf28}+1.88\%$
test_cql_speed[False-backward] 22.5387ms 22.0300ms 45.3926 Ops/s 44.0823 Ops/s $\color{#35bf28}+2.97\%$
test_cql_speed[True-None] 3.3734ms 3.2565ms 307.0755 Ops/s 288.0586 Ops/s $\textbf{\color{#35bf28}+6.60\%}$
test_cql_speed[True-backward] 5.9416ms 5.5068ms 181.5937 Ops/s 168.5248 Ops/s $\textbf{\color{#35bf28}+7.75\%}$
test_cql_speed[reduce-overhead-None] 19.5828ms 13.0608ms 76.5651 Ops/s 77.5248 Ops/s $\color{#d91a1a}-1.24\%$
test_cql_speed[reduce-overhead-backward] 2.1788ms 2.0126ms 496.8742 Ops/s 536.3137 Ops/s $\textbf{\color{#d91a1a}-7.35\%}$
test_a2c_speed[False-None] 3.3494ms 3.2140ms 311.1412 Ops/s 303.0811 Ops/s $\color{#35bf28}+2.66\%$
test_a2c_speed[False-backward] 6.9469ms 6.4478ms 155.0927 Ops/s 159.3577 Ops/s $\color{#d91a1a}-2.68\%$
test_a2c_speed[True-None] 1.7710ms 1.3770ms 726.2101 Ops/s 731.8989 Ops/s $\color{#d91a1a}-0.78\%$
test_a2c_speed[True-backward] 3.1096ms 3.0449ms 328.4139 Ops/s 335.4886 Ops/s $\color{#d91a1a}-2.11\%$
test_a2c_speed[reduce-overhead-None] 14.3449ms 8.4249ms 118.6963 Ops/s 119.0252 Ops/s $\color{#d91a1a}-0.28\%$
test_a2c_speed[reduce-overhead-backward] 1.7382ms 1.6181ms 618.0226 Ops/s 677.5779 Ops/s $\textbf{\color{#d91a1a}-8.79\%}$
test_ppo_speed[False-None] 3.9550ms 3.7362ms 267.6491 Ops/s 264.1491 Ops/s $\color{#35bf28}+1.33\%$
test_ppo_speed[False-backward] 7.5687ms 7.2320ms 138.2749 Ops/s 143.5088 Ops/s $\color{#d91a1a}-3.65\%$
test_ppo_speed[True-None] 1.4971ms 1.4148ms 706.8346 Ops/s 690.9732 Ops/s $\color{#35bf28}+2.30\%$
test_ppo_speed[True-backward] 3.3227ms 3.1887ms 313.6043 Ops/s 314.1781 Ops/s $\color{#d91a1a}-0.18\%$
test_ppo_speed[reduce-overhead-None] 1.0548ms 0.9648ms 1.0365 KOps/s 1.0272 KOps/s $\color{#35bf28}+0.91\%$
test_ppo_speed[reduce-overhead-backward] 1.7551ms 1.5700ms 636.9557 Ops/s 636.6684 Ops/s $\color{#35bf28}+0.05\%$
test_reinforce_speed[False-None] 2.4919ms 2.3067ms 433.5148 Ops/s 431.1993 Ops/s $\color{#35bf28}+0.54\%$
test_reinforce_speed[False-backward] 3.8819ms 3.4293ms 291.6046 Ops/s 285.3113 Ops/s $\color{#35bf28}+2.21\%$
test_reinforce_speed[True-None] 1.3781ms 1.2936ms 773.0197 Ops/s 741.8847 Ops/s $\color{#35bf28}+4.20\%$
test_reinforce_speed[True-backward] 3.2190ms 3.0565ms 327.1669 Ops/s 332.0961 Ops/s $\color{#d91a1a}-1.48\%$
test_reinforce_speed[reduce-overhead-None] 16.3455ms 9.3233ms 107.2579 Ops/s 108.3678 Ops/s $\color{#d91a1a}-1.02\%$
test_reinforce_speed[reduce-overhead-backward] 1.7596ms 1.6488ms 606.5000 Ops/s 642.9130 Ops/s $\textbf{\color{#d91a1a}-5.66\%}$
test_iql_speed[False-None] 9.6830ms 9.2432ms 108.1877 Ops/s 105.8223 Ops/s $\color{#35bf28}+2.24\%$
test_iql_speed[False-backward] 13.6948ms 13.2394ms 75.5320 Ops/s 75.1343 Ops/s $\color{#35bf28}+0.53\%$
test_iql_speed[True-None] 2.5381ms 2.2463ms 445.1690 Ops/s 436.4264 Ops/s $\color{#35bf28}+2.00\%$
test_iql_speed[True-backward] 5.2634ms 4.9215ms 203.1903 Ops/s 195.2441 Ops/s $\color{#35bf28}+4.07\%$
test_iql_speed[reduce-overhead-None] 0.4740s 12.6922ms 78.7886 Ops/s 95.2948 Ops/s $\textbf{\color{#d91a1a}-17.32\%}$
test_iql_speed[reduce-overhead-backward] 1.9509ms 1.9065ms 524.5157 Ops/s 479.3627 Ops/s $\textbf{\color{#35bf28}+9.42\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0214ms 6.3831ms 156.6639 Ops/s 154.3975 Ops/s $\color{#35bf28}+1.47\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5186ms 0.2635ms 3.7954 KOps/s 3.7772 KOps/s $\color{#35bf28}+0.48\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4674ms 0.2451ms 4.0793 KOps/s 4.1352 KOps/s $\color{#d91a1a}-1.35\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3895ms 6.1090ms 163.6923 Ops/s 162.5044 Ops/s $\color{#35bf28}+0.73\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9587ms 0.2598ms 3.8498 KOps/s 3.1632 KOps/s $\textbf{\color{#35bf28}+21.71\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4820ms 0.2415ms 4.1415 KOps/s 3.2782 KOps/s $\textbf{\color{#35bf28}+26.34\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4603ms 1.2712ms 786.6350 Ops/s 713.0684 Ops/s $\textbf{\color{#35bf28}+10.32\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4169ms 1.1682ms 856.0291 Ops/s 762.7915 Ops/s $\textbf{\color{#35bf28}+12.22\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5033ms 6.3227ms 158.1598 Ops/s 156.3996 Ops/s $\color{#35bf28}+1.13\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9337ms 0.4523ms 2.2112 KOps/s 2.0592 KOps/s $\textbf{\color{#35bf28}+7.38\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7833ms 0.4049ms 2.4695 KOps/s 2.2126 KOps/s $\textbf{\color{#35bf28}+11.61\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.3734ms 6.1465ms 162.6946 Ops/s 159.4855 Ops/s $\color{#35bf28}+2.01\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0631ms 0.3728ms 2.6824 KOps/s 3.0961 KOps/s $\textbf{\color{#d91a1a}-13.36\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 1.3001ms 0.3591ms 2.7847 KOps/s 4.1237 KOps/s $\textbf{\color{#d91a1a}-32.47\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 8.4318ms 6.0885ms 164.2451 Ops/s 161.7091 Ops/s $\color{#35bf28}+1.57\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9293ms 0.2729ms 3.6648 KOps/s 2.7968 KOps/s $\textbf{\color{#35bf28}+31.04\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5619ms 0.3263ms 3.0650 KOps/s 4.1443 KOps/s $\textbf{\color{#d91a1a}-26.04\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5395ms 6.2820ms 159.1862 Ops/s 156.5241 Ops/s $\color{#35bf28}+1.70\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8049ms 0.4158ms 2.4053 KOps/s 2.1902 KOps/s $\textbf{\color{#35bf28}+9.82\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.5780ms 0.3823ms 2.6158 KOps/s 2.4216 KOps/s $\textbf{\color{#35bf28}+8.02\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1195ms 5.4773ms 182.5708 Ops/s 179.4364 Ops/s $\color{#35bf28}+1.75\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 6.7127ms 2.0334ms 491.7861 Ops/s 443.9051 Ops/s $\textbf{\color{#35bf28}+10.79\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 9.0427ms 1.2971ms 770.9435 Ops/s 851.7121 Ops/s $\textbf{\color{#d91a1a}-9.48\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4519s 14.5424ms 68.7645 Ops/s 177.7569 Ops/s $\textbf{\color{#d91a1a}-61.32\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.1075ms 2.0523ms 487.2699 Ops/s 432.5327 Ops/s $\textbf{\color{#35bf28}+12.66\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.3137ms 1.3499ms 740.7872 Ops/s 843.2646 Ops/s $\textbf{\color{#d91a1a}-12.15\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.0515ms 5.8302ms 171.5197 Ops/s 31.2030 Ops/s $\textbf{\color{#35bf28}+449.69\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.8110ms 2.2831ms 438.0031 Ops/s 426.2722 Ops/s $\color{#35bf28}+2.75\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.6240ms 1.4927ms 669.9259 Ops/s 761.5669 Ops/s $\textbf{\color{#d91a1a}-12.03\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.6895ms 13.1994ms 75.7613 Ops/s 71.5810 Ops/s $\textbf{\color{#35bf28}+5.84\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.8552ms 17.1224ms 58.4032 Ops/s 58.5390 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 17.9801ms 17.5877ms 56.8578 Ops/s 54.3947 Ops/s $\color{#35bf28}+4.53\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7343ms 17.1906ms 58.1712 Ops/s 56.5773 Ops/s $\color{#35bf28}+2.82\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 17.9247ms 17.4892ms 57.1782 Ops/s 54.7467 Ops/s $\color{#35bf28}+4.44\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.8717ms 18.2035ms 54.9345 Ops/s 53.5243 Ops/s $\color{#35bf28}+2.63\%$

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.

3 participants