Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Deprecation] Deprecate default num_cells in MLP #2395

Merged
merged 2 commits into from
Aug 13, 2024
Merged

[Deprecation] Deprecate default num_cells in MLP #2395

merged 2 commits into from
Aug 13, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Aug 13, 2024 •
edited
Loading

closes #2328

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Aug 13, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2395

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

❌ 2 New Failures, 4 Unrelated Failures

As of commit da6400d with merge base bb0ddb5 (image):

NEW FAILURES - The following jobs have failed:

FLAKY - The following job failed but was likely due to flakiness present on trunk:

BROKEN TRUNK - The following jobs failed but was 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 Aug 13, 2024
@vmoens vmoens added the quality code quality label Aug 13, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Aug 13, 2024 •
edited
Loading

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

Total Benchmarks: 91. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}30$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 61.8279ms 60.6266ms 16.4944 Ops/s 17.1085 Ops/s $\color{#d91a1a}-3.59\%$
test_sync 39.1223ms 33.3943ms 29.9452 Ops/s 26.0562 Ops/s $\textbf{\color{#35bf28}+14.93\%}$
test_async 85.7451ms 31.5641ms 31.6815 Ops/s 32.4759 Ops/s $\color{#d91a1a}-2.45\%$
test_simple 0.5045s 0.4300s 2.3257 Ops/s 2.3873 Ops/s $\color{#d91a1a}-2.58\%$
test_transformed 0.6687s 0.5932s 1.6856 Ops/s 1.7435 Ops/s $\color{#d91a1a}-3.32\%$
test_serial 1.3878s 1.3096s 0.7636 Ops/s 0.7942 Ops/s $\color{#d91a1a}-3.86\%$
test_parallel 1.1156s 1.1037s 0.9060 Ops/s 0.8958 Ops/s $\color{#35bf28}+1.14\%$
test_step_mdp_speed[True-True-True-True-True] 0.2479ms 26.4152μs 37.8570 KOps/s 40.7213 KOps/s $\textbf{\color{#d91a1a}-7.03\%}$
test_step_mdp_speed[True-True-True-True-False] 65.2720μs 15.4327μs 64.7974 KOps/s 70.2031 KOps/s $\textbf{\color{#d91a1a}-7.70\%}$
test_step_mdp_speed[True-True-True-False-True] 46.7680μs 15.1317μs 66.0863 KOps/s 70.8689 KOps/s $\textbf{\color{#d91a1a}-6.75\%}$
test_step_mdp_speed[True-True-True-False-False] 63.3780μs 8.9365μs 111.9008 KOps/s 120.4185 KOps/s $\textbf{\color{#d91a1a}-7.07\%}$
test_step_mdp_speed[True-True-False-True-True] 62.0260μs 27.7969μs 35.9752 KOps/s 37.8517 KOps/s $\color{#d91a1a}-4.96\%$
test_step_mdp_speed[True-True-False-True-False] 44.4440μs 17.0584μs 58.6223 KOps/s 63.2538 KOps/s $\textbf{\color{#d91a1a}-7.32\%}$
test_step_mdp_speed[True-True-False-False-True] 41.3680μs 16.8242μs 59.4383 KOps/s 63.8854 KOps/s $\textbf{\color{#d91a1a}-6.96\%}$
test_step_mdp_speed[True-True-False-False-False] 59.4110μs 10.4772μs 95.4457 KOps/s 102.6715 KOps/s $\textbf{\color{#d91a1a}-7.04\%}$
test_step_mdp_speed[True-False-True-True-True] 80.3900μs 29.6418μs 33.7361 KOps/s 35.6253 KOps/s $\textbf{\color{#d91a1a}-5.30\%}$
test_step_mdp_speed[True-False-True-True-False] 48.2100μs 18.7922μs 53.2136 KOps/s 57.0416 KOps/s $\textbf{\color{#d91a1a}-6.71\%}$
test_step_mdp_speed[True-False-True-False-True] 65.2520μs 16.7310μs 59.7692 KOps/s 63.1676 KOps/s $\textbf{\color{#d91a1a}-5.38\%}$
test_step_mdp_speed[True-False-True-False-False] 61.1050μs 10.6058μs 94.2876 KOps/s 102.4772 KOps/s $\textbf{\color{#d91a1a}-7.99\%}$
test_step_mdp_speed[True-False-False-True-True] 86.4320μs 31.0425μs 32.2139 KOps/s 34.4388 KOps/s $\textbf{\color{#d91a1a}-6.46\%}$
test_step_mdp_speed[True-False-False-True-False] 75.9320μs 20.3726μs 49.0855 KOps/s 53.5612 KOps/s $\textbf{\color{#d91a1a}-8.36\%}$
test_step_mdp_speed[True-False-False-False-True] 47.8200μs 18.1703μs 55.0349 KOps/s 58.2568 KOps/s $\textbf{\color{#d91a1a}-5.53\%}$
test_step_mdp_speed[True-False-False-False-False] 58.9800μs 12.1663μs 82.1943 KOps/s 90.1878 KOps/s $\textbf{\color{#d91a1a}-8.86\%}$
test_step_mdp_speed[False-True-True-True-True] 60.3640μs 29.8392μs 33.5129 KOps/s 35.3471 KOps/s $\textbf{\color{#d91a1a}-5.19\%}$
test_step_mdp_speed[False-True-True-True-False] 64.0700μs 18.7866μs 53.2295 KOps/s 54.3147 KOps/s $\color{#d91a1a}-2.00\%$
test_step_mdp_speed[False-True-True-False-True] 50.7250μs 19.2316μs 51.9978 KOps/s 52.4322 KOps/s $\color{#d91a1a}-0.83\%$
test_step_mdp_speed[False-True-True-False-False] 42.1890μs 11.7331μs 85.2290 KOps/s 90.0293 KOps/s $\textbf{\color{#d91a1a}-5.33\%}$
test_step_mdp_speed[False-True-False-True-True] 85.1290μs 31.0716μs 32.1837 KOps/s 33.4795 KOps/s $\color{#d91a1a}-3.87\%$
test_step_mdp_speed[False-True-False-True-False] 49.3320μs 20.3552μs 49.1275 KOps/s 52.7324 KOps/s $\textbf{\color{#d91a1a}-6.84\%}$
test_step_mdp_speed[False-True-False-False-True] 77.9160μs 20.7537μs 48.1842 KOps/s 49.8135 KOps/s $\color{#d91a1a}-3.27\%$
test_step_mdp_speed[False-True-False-False-False] 40.9470μs 13.4493μs 74.3535 KOps/s 80.3138 KOps/s $\textbf{\color{#d91a1a}-7.42\%}$
test_step_mdp_speed[False-False-True-True-True] 3.8266ms 32.6426μs 30.6349 KOps/s 31.6976 KOps/s $\color{#d91a1a}-3.35\%$
test_step_mdp_speed[False-False-True-True-False] 0.1141ms 22.5740μs 44.2987 KOps/s 48.4525 KOps/s $\textbf{\color{#d91a1a}-8.57\%}$
test_step_mdp_speed[False-False-True-False-True] 54.7030μs 20.9927μs 47.6356 KOps/s 50.3783 KOps/s $\textbf{\color{#d91a1a}-5.44\%}$
test_step_mdp_speed[False-False-True-False-False] 60.7640μs 13.4181μs 74.5263 KOps/s 80.2581 KOps/s $\textbf{\color{#d91a1a}-7.14\%}$
test_step_mdp_speed[False-False-False-True-True] 0.1169ms 33.9291μs 29.4732 KOps/s 30.5871 KOps/s $\color{#d91a1a}-3.64\%$
test_step_mdp_speed[False-False-False-True-False] 55.0030μs 23.5128μs 42.5301 KOps/s 45.3243 KOps/s $\textbf{\color{#d91a1a}-6.16\%}$
test_step_mdp_speed[False-False-False-False-True] 78.0460μs 22.1124μs 45.2235 KOps/s 47.1059 KOps/s $\color{#d91a1a}-4.00\%$
test_step_mdp_speed[False-False-False-False-False] 67.5960μs 14.8548μs 67.3185 KOps/s 72.2184 KOps/s $\textbf{\color{#d91a1a}-6.78\%}$
test_values[generalized_advantage_estimate-True-True] 11.2111ms 9.7667ms 102.3884 Ops/s 103.9553 Ops/s $\color{#d91a1a}-1.51\%$
test_values[vec_generalized_advantage_estimate-True-True] 40.8119ms 36.1384ms 27.6714 Ops/s 27.8283 Ops/s $\color{#d91a1a}-0.56\%$
test_values[td0_return_estimate-False-False] 0.2459ms 0.1844ms 5.4243 KOps/s 5.9437 KOps/s $\textbf{\color{#d91a1a}-8.74\%}$
test_values[td1_return_estimate-False-False] 23.9693ms 23.7300ms 42.1407 Ops/s 41.9098 Ops/s $\color{#35bf28}+0.55\%$
test_values[vec_td1_return_estimate-False-False] 38.1229ms 35.9434ms 27.8215 Ops/s 27.6630 Ops/s $\color{#35bf28}+0.57\%$
test_values[td_lambda_return_estimate-True-False] 34.7803ms 34.5463ms 28.9467 Ops/s 28.7462 Ops/s $\color{#35bf28}+0.70\%$
test_values[vec_td_lambda_return_estimate-True-False] 37.7092ms 36.0236ms 27.7596 Ops/s 27.7848 Ops/s $\color{#d91a1a}-0.09\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.2779ms 8.1543ms 122.6351 Ops/s 119.9209 Ops/s $\color{#35bf28}+2.26\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8834ms 1.7805ms 561.6519 Ops/s 509.0561 Ops/s $\textbf{\color{#35bf28}+10.33\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6192ms 0.3621ms 2.7613 KOps/s 2.7512 KOps/s $\color{#35bf28}+0.37\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 50.1390ms 48.2209ms 20.7379 Ops/s 20.9422 Ops/s $\color{#d91a1a}-0.98\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.2004ms 3.0368ms 329.2978 Ops/s 327.7997 Ops/s $\color{#35bf28}+0.46\%$
test_dqn_speed 1.7456ms 1.3071ms 765.0752 Ops/s 762.4321 Ops/s $\color{#35bf28}+0.35\%$
test_ddpg_speed 3.5288ms 2.7389ms 365.1147 Ops/s 367.6112 Ops/s $\color{#d91a1a}-0.68\%$
test_sac_speed 9.3364ms 8.1126ms 123.2657 Ops/s 124.5386 Ops/s $\color{#d91a1a}-1.02\%$
test_redq_speed 13.9739ms 13.0518ms 76.6175 Ops/s 76.8827 Ops/s $\color{#d91a1a}-0.34\%$
test_redq_deprec_speed 14.5440ms 13.1245ms 76.1932 Ops/s 76.9292 Ops/s $\color{#d91a1a}-0.96\%$
test_td3_speed 9.8387ms 8.0734ms 123.8635 Ops/s 124.6105 Ops/s $\color{#d91a1a}-0.60\%$
test_cql_speed 38.4804ms 36.1113ms 27.6922 Ops/s 25.6179 Ops/s $\textbf{\color{#35bf28}+8.10\%}$
test_a2c_speed 8.3015ms 7.4604ms 134.0417 Ops/s 136.5929 Ops/s $\color{#d91a1a}-1.87\%$
test_ppo_speed 9.7984ms 7.8122ms 128.0049 Ops/s 131.6722 Ops/s $\color{#d91a1a}-2.79\%$
test_reinforce_speed 7.3490ms 6.4806ms 154.3071 Ops/s 155.6597 Ops/s $\color{#d91a1a}-0.87\%$
test_iql_speed 38.2519ms 32.8901ms 30.4043 Ops/s 31.0613 Ops/s $\color{#d91a1a}-2.12\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8263ms 4.9825ms 200.7037 Ops/s 203.2041 Ops/s $\color{#d91a1a}-1.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8418ms 0.4908ms 2.0375 KOps/s 2.0741 KOps/s $\color{#d91a1a}-1.76\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6387ms 0.4610ms 2.1694 KOps/s 2.1860 KOps/s $\color{#d91a1a}-0.76\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.0474ms 4.9937ms 200.2514 Ops/s 207.0957 Ops/s $\color{#d91a1a}-3.30\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.3549ms 0.4802ms 2.0825 KOps/s 2.1085 KOps/s $\color{#d91a1a}-1.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9622ms 0.4589ms 2.1789 KOps/s 2.1997 KOps/s $\color{#d91a1a}-0.94\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.4796ms 1.7101ms 584.7551 Ops/s 591.4822 Ops/s $\color{#d91a1a}-1.14\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.2459ms 1.6142ms 619.5037 Ops/s 618.7536 Ops/s $\color{#35bf28}+0.12\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.3858ms 5.1394ms 194.5737 Ops/s 198.9126 Ops/s $\color{#d91a1a}-2.18\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8640ms 0.6254ms 1.5989 KOps/s 1.6065 KOps/s $\color{#d91a1a}-0.47\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.2484ms 0.6097ms 1.6400 KOps/s 1.6940 KOps/s $\color{#d91a1a}-3.19\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.1316ms 4.9152ms 203.4488 Ops/s 206.8646 Ops/s $\color{#d91a1a}-1.65\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7028ms 0.4859ms 2.0580 KOps/s 2.0934 KOps/s $\color{#d91a1a}-1.69\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 7.1352ms 0.4695ms 2.1298 KOps/s 2.1873 KOps/s $\color{#d91a1a}-2.63\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.6993ms 5.0321ms 198.7250 Ops/s 209.9596 Ops/s $\textbf{\color{#d91a1a}-5.35\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9148ms 0.4854ms 2.0602 KOps/s 2.0781 KOps/s $\color{#d91a1a}-0.86\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6442ms 0.4600ms 2.1739 KOps/s 2.1975 KOps/s $\color{#d91a1a}-1.07\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.9980ms 5.0536ms 197.8788 Ops/s 199.9243 Ops/s $\color{#d91a1a}-1.02\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.5603ms 0.6221ms 1.6076 KOps/s 1.5983 KOps/s $\color{#35bf28}+0.58\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7649ms 0.5992ms 1.6689 KOps/s 1.6985 KOps/s $\color{#d91a1a}-1.74\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1332s 8.9435ms 111.8136 Ops/s 158.2312 Ops/s $\textbf{\color{#d91a1a}-29.34\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 17.9359ms 13.0595ms 76.5727 Ops/s 77.1581 Ops/s $\color{#d91a1a}-0.76\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.9269ms 1.1228ms 890.6550 Ops/s 920.1842 Ops/s $\color{#d91a1a}-3.21\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1196s 6.0217ms 166.0663 Ops/s 168.2107 Ops/s $\color{#d91a1a}-1.27\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 16.7890ms 12.8809ms 77.6341 Ops/s 65.7264 Ops/s $\textbf{\color{#35bf28}+18.12\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 4.6803ms 1.1760ms 850.3143 Ops/s 935.5284 Ops/s $\textbf{\color{#d91a1a}-9.11\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1125s 6.0310ms 165.8099 Ops/s 160.0404 Ops/s $\color{#35bf28}+3.61\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.1240s 15.2818ms 65.4374 Ops/s 75.8736 Ops/s $\textbf{\color{#d91a1a}-13.75\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 4.9154ms 1.3621ms 734.1595 Ops/s 817.9534 Ops/s $\textbf{\color{#d91a1a}-10.24\%}$

@github-actions GitHub Actions
Copy link

github-actions bot commented Aug 13, 2024 •
edited
Loading

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

Total Benchmarks: 94. Improved: $\large\color{#35bf28}1$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1065s 0.1063s 9.4114 Ops/s 9.4501 Ops/s $\color{#d91a1a}-0.41\%$
test_sync 95.2845ms 91.8440ms 10.8880 Ops/s 10.8692 Ops/s $\color{#35bf28}+0.17\%$
test_async 0.1789s 89.0733ms 11.2267 Ops/s 11.2949 Ops/s $\color{#d91a1a}-0.60\%$
test_single_pixels 0.1182s 0.1179s 8.4814 Ops/s 8.6343 Ops/s $\color{#d91a1a}-1.77\%$
test_sync_pixels 77.4641ms 75.7245ms 13.2058 Ops/s 12.9410 Ops/s $\color{#35bf28}+2.05\%$
test_async_pixels 0.1320s 70.4993ms 14.1845 Ops/s 14.2105 Ops/s $\color{#d91a1a}-0.18\%$
test_simple 0.7560s 0.7552s 1.3242 Ops/s 1.2633 Ops/s $\color{#35bf28}+4.82\%$
test_transformed 1.0882s 1.0100s 0.9901 Ops/s 1.0146 Ops/s $\color{#d91a1a}-2.42\%$
test_serial 2.2372s 2.1636s 0.4622 Ops/s 0.4706 Ops/s $\color{#d91a1a}-1.79\%$
test_parallel 1.9643s 1.8936s 0.5281 Ops/s 0.5285 Ops/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[True-True-True-True-True] 0.1791ms 38.0637μs 26.2717 KOps/s 25.8356 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[True-True-True-True-False] 0.1673ms 21.3802μs 46.7722 KOps/s 46.3647 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[True-True-True-False-True] 0.1471ms 20.7676μs 48.1519 KOps/s 45.9430 KOps/s $\color{#35bf28}+4.81\%$
test_step_mdp_speed[True-True-True-False-False] 35.2820μs 11.8792μs 84.1808 KOps/s 82.0232 KOps/s $\color{#35bf28}+2.63\%$
test_step_mdp_speed[True-True-False-True-True] 0.2063ms 40.3441μs 24.7868 KOps/s 24.7034 KOps/s $\color{#35bf28}+0.34\%$
test_step_mdp_speed[True-True-False-True-False] 40.2720μs 23.5330μs 42.4935 KOps/s 42.3180 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-True-False-False-True] 43.4530μs 23.7457μs 42.1129 KOps/s 41.1761 KOps/s $\color{#35bf28}+2.28\%$
test_step_mdp_speed[True-True-False-False-False] 30.5920μs 14.3444μs 69.7138 KOps/s 69.9556 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-False-True-True-True] 62.9840μs 42.6887μs 23.4254 KOps/s 23.2998 KOps/s $\color{#35bf28}+0.54\%$
test_step_mdp_speed[True-False-True-True-False] 88.4160μs 25.8056μs 38.7512 KOps/s 38.8777 KOps/s $\color{#d91a1a}-0.33\%$
test_step_mdp_speed[True-False-True-False-True] 40.8020μs 23.4741μs 42.6002 KOps/s 41.6247 KOps/s $\color{#35bf28}+2.34\%$
test_step_mdp_speed[True-False-True-False-False] 0.2046ms 14.1653μs 70.5953 KOps/s 69.4894 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-False-False-True-True] 94.8460μs 44.9338μs 22.2549 KOps/s 22.1836 KOps/s $\color{#35bf28}+0.32\%$
test_step_mdp_speed[True-False-False-True-False] 44.7930μs 28.2036μs 35.4565 KOps/s 36.1278 KOps/s $\color{#d91a1a}-1.86\%$
test_step_mdp_speed[True-False-False-False-True] 99.9960μs 25.1929μs 39.6937 KOps/s 37.9669 KOps/s $\color{#35bf28}+4.55\%$
test_step_mdp_speed[True-False-False-False-False] 0.2130ms 16.4365μs 60.8403 KOps/s 61.0872 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[False-True-True-True-True] 59.2540μs 42.8927μs 23.3140 KOps/s 23.6543 KOps/s $\color{#d91a1a}-1.44\%$
test_step_mdp_speed[False-True-True-True-False] 0.2067ms 25.9341μs 38.5593 KOps/s 38.7306 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[False-True-True-False-True] 50.3330μs 28.4060μs 35.2038 KOps/s 35.0280 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[False-True-True-False-False] 30.4610μs 16.0214μs 62.4165 KOps/s 61.4735 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-True-False-True-True] 71.7140μs 44.6187μs 22.4121 KOps/s 22.3698 KOps/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[False-True-False-True-False] 46.1330μs 27.9166μs 35.8210 KOps/s 36.1038 KOps/s $\color{#d91a1a}-0.78\%$
test_step_mdp_speed[False-True-False-False-True] 48.2930μs 30.2004μs 33.1122 KOps/s 33.0174 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[False-True-False-False-False] 59.9240μs 18.1713μs 55.0319 KOps/s 54.0411 KOps/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[False-False-True-True-True] 4.0361ms 47.5226μs 21.0426 KOps/s 20.8438 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[False-False-True-True-False] 48.5130μs 30.5514μs 32.7317 KOps/s 32.8146 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[False-False-True-False-True] 0.1096ms 29.8408μs 33.5112 KOps/s 32.5403 KOps/s $\color{#35bf28}+2.98\%$
test_step_mdp_speed[False-False-True-False-False] 38.8920μs 18.1995μs 54.9465 KOps/s 52.5531 KOps/s $\color{#35bf28}+4.55\%$
test_step_mdp_speed[False-False-False-True-True] 0.1273ms 47.9476μs 20.8561 KOps/s 20.1839 KOps/s $\color{#35bf28}+3.33\%$
test_step_mdp_speed[False-False-False-True-False] 0.2175ms 32.3628μs 30.8997 KOps/s 30.7618 KOps/s $\color{#35bf28}+0.45\%$
test_step_mdp_speed[False-False-False-False-True] 49.0730μs 31.6370μs 31.6086 KOps/s 30.9979 KOps/s $\color{#35bf28}+1.97\%$
test_step_mdp_speed[False-False-False-False-False] 43.7420μs 20.2946μs 49.2741 KOps/s 47.9376 KOps/s $\color{#35bf28}+2.79\%$
test_values[generalized_advantage_estimate-True-True] 25.1408ms 24.0660ms 41.5524 Ops/s 41.7176 Ops/s $\color{#d91a1a}-0.40\%$
test_values[vec_generalized_advantage_estimate-True-True] 90.8617ms 2.7183ms 367.8719 Ops/s 365.3001 Ops/s $\color{#35bf28}+0.70\%$
test_values[td0_return_estimate-False-False] 91.7050μs 65.8961μs 15.1754 KOps/s 15.1240 KOps/s $\color{#35bf28}+0.34\%$
test_values[td1_return_estimate-False-False] 57.7069ms 55.0633ms 18.1609 Ops/s 18.3245 Ops/s $\color{#d91a1a}-0.89\%$
test_values[vec_td1_return_estimate-False-False] 1.4632ms 1.0882ms 918.9372 Ops/s 923.2565 Ops/s $\color{#d91a1a}-0.47\%$
test_values[td_lambda_return_estimate-True-False] 87.4843ms 87.0532ms 11.4872 Ops/s 11.0964 Ops/s $\color{#35bf28}+3.52\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4304ms 1.0861ms 920.7334 Ops/s 922.0274 Ops/s $\color{#d91a1a}-0.14\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 24.7161ms 23.8732ms 41.8880 Ops/s 41.6018 Ops/s $\color{#35bf28}+0.69\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.9913ms 0.7478ms 1.3372 KOps/s 1.3894 KOps/s $\color{#d91a1a}-3.75\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8250ms 0.6674ms 1.4983 KOps/s 1.4943 KOps/s $\color{#35bf28}+0.27\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6237ms 1.4708ms 679.9112 Ops/s 681.5837 Ops/s $\color{#d91a1a}-0.25\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8560ms 0.6846ms 1.4608 KOps/s 1.4616 KOps/s $\color{#d91a1a}-0.06\%$
test_dqn_speed 7.7906ms 1.3614ms 734.5121 Ops/s 732.5367 Ops/s $\color{#35bf28}+0.27\%$
test_ddpg_speed 3.0121ms 2.7568ms 362.7360 Ops/s 362.9060 Ops/s $\color{#d91a1a}-0.05\%$
test_sac_speed 8.3195ms 7.9862ms 125.2161 Ops/s 125.8511 Ops/s $\color{#d91a1a}-0.50\%$
test_redq_speed 12.9025ms 10.4087ms 96.0738 Ops/s 95.6899 Ops/s $\color{#35bf28}+0.40\%$
test_redq_deprec_speed 11.4597ms 11.0800ms 90.2528 Ops/s 89.8506 Ops/s $\color{#35bf28}+0.45\%$
test_td3_speed 7.9994ms 7.8955ms 126.6547 Ops/s 126.7995 Ops/s $\color{#d91a1a}-0.11\%$
test_cql_speed 26.3019ms 25.5716ms 39.1058 Ops/s 38.0452 Ops/s $\color{#35bf28}+2.79\%$
test_a2c_speed 5.8390ms 5.5814ms 179.1668 Ops/s 176.1997 Ops/s $\color{#35bf28}+1.68\%$
test_ppo_speed 6.2722ms 5.9393ms 168.3701 Ops/s 167.8199 Ops/s $\color{#35bf28}+0.33\%$
test_reinforce_speed 4.8209ms 4.5492ms 219.8210 Ops/s 219.1764 Ops/s $\color{#35bf28}+0.29\%$
test_iql_speed 20.8467ms 19.5681ms 51.1036 Ops/s 50.4579 Ops/s $\color{#35bf28}+1.28\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.8183ms 6.6544ms 150.2763 Ops/s 151.5705 Ops/s $\color{#d91a1a}-0.85\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8687ms 0.5293ms 1.8892 KOps/s 1.8972 KOps/s $\color{#d91a1a}-0.42\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7072ms 0.5102ms 1.9601 KOps/s 1.9711 KOps/s $\color{#d91a1a}-0.56\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.9440ms 6.5525ms 152.6143 Ops/s 153.3893 Ops/s $\color{#d91a1a}-0.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.5813ms 0.5232ms 1.9114 KOps/s 1.9107 KOps/s $\color{#35bf28}+0.03\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6963ms 0.5042ms 1.9832 KOps/s 1.9716 KOps/s $\color{#35bf28}+0.59\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.2550ms 2.0273ms 493.2787 Ops/s 504.2545 Ops/s $\color{#d91a1a}-2.18\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.1328ms 1.9158ms 521.9625 Ops/s 530.3443 Ops/s $\color{#d91a1a}-1.58\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.0074ms 6.7799ms 147.4956 Ops/s 148.5789 Ops/s $\color{#d91a1a}-0.73\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.3875ms 0.7113ms 1.4059 KOps/s 1.4787 KOps/s $\color{#d91a1a}-4.92\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8167ms 0.6648ms 1.5041 KOps/s 1.5245 KOps/s $\color{#d91a1a}-1.33\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.9134ms 6.6331ms 150.7593 Ops/s 151.7374 Ops/s $\color{#d91a1a}-0.64\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7127ms 0.5271ms 1.8971 KOps/s 1.8948 KOps/s $\color{#35bf28}+0.12\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.1606ms 0.5149ms 1.9421 KOps/s 1.9616 KOps/s $\color{#d91a1a}-0.99\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.9072ms 6.5507ms 152.6544 Ops/s 153.1870 Ops/s $\color{#d91a1a}-0.35\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.1270s 0.6951ms 1.4386 KOps/s 1.9198 KOps/s $\textbf{\color{#d91a1a}-25.06\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6764ms 0.4952ms 2.0194 KOps/s 1.9736 KOps/s $\color{#35bf28}+2.32\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.9519ms 6.7764ms 147.5714 Ops/s 148.2329 Ops/s $\color{#d91a1a}-0.45\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.0814ms 0.6851ms 1.4597 KOps/s 1.4751 KOps/s $\color{#d91a1a}-1.04\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8341ms 0.6602ms 1.5147 KOps/s 1.5130 KOps/s $\color{#35bf28}+0.11\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1378s 7.8473ms 127.4331 Ops/s 124.0025 Ops/s $\color{#35bf28}+2.77\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 18.3388ms 15.9661ms 62.6326 Ops/s 54.5096 Ops/s $\textbf{\color{#35bf28}+14.90\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.5899ms 1.3615ms 734.4613 Ops/s 857.9450 Ops/s $\textbf{\color{#d91a1a}-14.39\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1284s 7.6455ms 130.7956 Ops/s 128.6631 Ops/s $\color{#35bf28}+1.66\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1439s 18.4907ms 54.0813 Ops/s 63.4079 Ops/s $\textbf{\color{#d91a1a}-14.71\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.9096ms 1.3527ms 739.2634 Ops/s 751.9864 Ops/s $\color{#d91a1a}-1.69\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1294s 7.8687ms 127.0855 Ops/s 125.8139 Ops/s $\color{#35bf28}+1.01\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 18.6025ms 16.1680ms 61.8505 Ops/s 62.2556 Ops/s $\color{#d91a1a}-0.65\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.4543ms 1.4349ms 696.9131 Ops/s 691.8617 Ops/s $\color{#35bf28}+0.73\%$

@vmoens vmoens merged commit 25e8bd2 into main Aug 13, 2024
69 of 71 checks passed
@vmoens vmoens deleted the fix-mlp-default branch August 13, 2024 20:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. quality code quality
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[BUG] Surprising MLP default behavior
2 participants
@vmoens @facebook-github-bot