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

Revert "[Refactor] Deprecate direct usage of memmap tensors" #1698

Merged
merged 1 commit into from
Nov 15, 2023
Merged

Revert "[Refactor] Deprecate direct usage of memmap tensors" #1698

merged 1 commit into from
Nov 15, 2023

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Nov 15, 2023

Reverts #1684

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Nov 15, 2023 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1698

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

❌ 4 New Failures, 23 Unrelated Failures

As of commit 584bfa3 with merge base 0badd6e (image):

NEW FAILURES - The following jobs have failed:

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

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@vmoens vmoens marked this pull request as ready for review November 15, 2023 13:54
@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 Nov 15, 2023
@vmoens vmoens merged commit d82aa45 into main Nov 15, 2023
11 of 27 checks passed
@vmoens vmoens deleted the revert-1684-deprecate_direct_memmap branch November 15, 2023 13:54
@github-actions GitHub Actions
Copy link

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 62.6042ms 61.6901ms 16.2101 Ops/s 16.0001 Ops/s $\color{#35bf28}+1.31\%$
test_sync 34.7144ms 33.9845ms 29.4252 Ops/s 29.8718 Ops/s $\color{#d91a1a}-1.50\%$
test_async 63.8774ms 31.1124ms 32.1415 Ops/s 30.1104 Ops/s $\textbf{\color{#35bf28}+6.75\%}$
test_simple 0.4998s 0.4445s 2.2497 Ops/s 2.3335 Ops/s $\color{#d91a1a}-3.59\%$
test_transformed 0.6595s 0.6015s 1.6625 Ops/s 1.7025 Ops/s $\color{#d91a1a}-2.35\%$
test_serial 1.3933s 1.3073s 0.7649 Ops/s 0.8058 Ops/s $\textbf{\color{#d91a1a}-5.08\%}$
test_parallel 1.2871s 1.2391s 0.8070 Ops/s 0.8166 Ops/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-True-True-True-True] 0.1698ms 24.8876μs 40.1807 KOps/s 40.7728 KOps/s $\color{#d91a1a}-1.45\%$
test_step_mdp_speed[True-True-True-True-False] 44.7040μs 14.8125μs 67.5108 KOps/s 68.2788 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[True-True-True-False-True] 47.8100μs 15.4363μs 64.7822 KOps/s 65.0834 KOps/s $\color{#d91a1a}-0.46\%$
test_step_mdp_speed[True-True-True-False-False] 45.3850μs 8.9720μs 111.4576 KOps/s 110.3913 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-True-False-True-True] 82.9980μs 26.4115μs 37.8623 KOps/s 38.7043 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[True-True-False-True-False] 41.6180μs 16.0439μs 62.3289 KOps/s 62.3668 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[True-True-False-False-True] 47.0190μs 16.7251μs 59.7902 KOps/s 60.5128 KOps/s $\color{#d91a1a}-1.19\%$
test_step_mdp_speed[True-True-False-False-False] 48.6110μs 10.3087μs 97.0056 KOps/s 95.9310 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[True-False-True-True-True] 84.8080μs 27.8561μs 35.8988 KOps/s 36.7995 KOps/s $\color{#d91a1a}-2.45\%$
test_step_mdp_speed[True-False-True-True-False] 66.8550μs 17.5472μs 56.9891 KOps/s 57.1355 KOps/s $\color{#d91a1a}-0.26\%$
test_step_mdp_speed[True-False-True-False-True] 2.2434ms 16.8767μs 59.2535 KOps/s 60.7895 KOps/s $\color{#d91a1a}-2.53\%$
test_step_mdp_speed[True-False-True-False-False] 35.8270μs 10.3176μs 96.9214 KOps/s 97.2849 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-False-False-True-True] 94.0170μs 28.9689μs 34.5197 KOps/s 35.5837 KOps/s $\color{#d91a1a}-2.99\%$
test_step_mdp_speed[True-False-False-True-False] 55.1840μs 18.9899μs 52.6595 KOps/s 53.5438 KOps/s $\color{#d91a1a}-1.65\%$
test_step_mdp_speed[True-False-False-False-True] 57.0880μs 17.8739μs 55.9476 KOps/s 56.7757 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[True-False-False-False-False] 41.7090μs 11.5639μs 86.4757 KOps/s 86.6404 KOps/s $\color{#d91a1a}-0.19\%$
test_step_mdp_speed[False-True-True-True-True] 63.2590μs 27.6227μs 36.2022 KOps/s 36.6050 KOps/s $\color{#d91a1a}-1.10\%$
test_step_mdp_speed[False-True-True-True-False] 70.8030μs 17.5996μs 56.8194 KOps/s 57.1625 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[False-True-True-False-True] 64.7020μs 19.4006μs 51.5447 KOps/s 52.8660 KOps/s $\color{#d91a1a}-2.50\%$
test_step_mdp_speed[False-True-True-False-False] 44.3140μs 11.8295μs 84.5343 KOps/s 86.6888 KOps/s $\color{#d91a1a}-2.49\%$
test_step_mdp_speed[False-True-False-True-True] 59.2320μs 28.9220μs 34.5758 KOps/s 34.7681 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-True-False-True-False] 59.0410μs 18.8801μs 52.9659 KOps/s 53.7560 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-True-False-False-True] 63.1890μs 20.4957μs 48.7908 KOps/s 49.1250 KOps/s $\color{#d91a1a}-0.68\%$
test_step_mdp_speed[False-True-False-False-False] 51.4670μs 12.9336μs 77.3179 KOps/s 76.4435 KOps/s $\color{#35bf28}+1.14\%$
test_step_mdp_speed[False-False-True-True-True] 85.9620μs 30.6670μs 32.6084 KOps/s 33.4132 KOps/s $\color{#d91a1a}-2.41\%$
test_step_mdp_speed[False-False-True-True-False] 58.4200μs 20.4693μs 48.8536 KOps/s 50.0775 KOps/s $\color{#d91a1a}-2.44\%$
test_step_mdp_speed[False-False-True-False-True] 59.3220μs 20.6069μs 48.5275 KOps/s 49.9282 KOps/s $\color{#d91a1a}-2.81\%$
test_step_mdp_speed[False-False-True-False-False] 46.6980μs 13.2499μs 75.4723 KOps/s 77.1579 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[False-False-False-True-True] 72.3860μs 31.1655μs 32.0868 KOps/s 32.8812 KOps/s $\color{#d91a1a}-2.42\%$
test_step_mdp_speed[False-False-False-True-False] 70.6820μs 21.3334μs 46.8749 KOps/s 47.1492 KOps/s $\color{#d91a1a}-0.58\%$
test_step_mdp_speed[False-False-False-False-True] 72.8450μs 21.1365μs 47.3116 KOps/s 47.0432 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[False-False-False-False-False] 38.2520μs 14.1661μs 70.5913 KOps/s 71.5732 KOps/s $\color{#d91a1a}-1.37\%$
test_values[generalized_advantage_estimate-True-True] 14.5412ms 12.0794ms 82.7853 Ops/s 84.5295 Ops/s $\color{#d91a1a}-2.06\%$
test_values[vec_generalized_advantage_estimate-True-True] 34.3767ms 26.6511ms 37.5220 Ops/s 37.5849 Ops/s $\color{#d91a1a}-0.17\%$
test_values[td0_return_estimate-False-False] 0.2321ms 0.2012ms 4.9706 KOps/s 5.6446 KOps/s $\textbf{\color{#d91a1a}-11.94\%}$
test_values[td1_return_estimate-False-False] 26.1389ms 25.7952ms 38.7669 Ops/s 40.2254 Ops/s $\color{#d91a1a}-3.63\%$
test_values[vec_td1_return_estimate-False-False] 34.7536ms 26.8395ms 37.2585 Ops/s 37.4322 Ops/s $\color{#d91a1a}-0.46\%$
test_values[td_lambda_return_estimate-True-False] 38.2664ms 36.4485ms 27.4360 Ops/s 28.2979 Ops/s $\color{#d91a1a}-3.05\%$
test_values[vec_td_lambda_return_estimate-True-False] 35.0445ms 26.5922ms 37.6050 Ops/s 36.9592 Ops/s $\color{#35bf28}+1.75\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.2246ms 8.0254ms 124.6040 Ops/s 124.8911 Ops/s $\color{#d91a1a}-0.23\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 10.7381ms 1.9455ms 513.9989 Ops/s 510.1685 Ops/s $\color{#35bf28}+0.75\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 1.4959ms 0.4346ms 2.3008 KOps/s 2.3180 KOps/s $\color{#d91a1a}-0.74\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 0.1718s 48.4928ms 20.6216 Ops/s 24.0536 Ops/s $\textbf{\color{#d91a1a}-14.27\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 10.9979ms 2.5449ms 392.9383 Ops/s 396.4152 Ops/s $\color{#d91a1a}-0.88\%$
test_dqn_speed 10.2434ms 1.7106ms 584.5752 Ops/s 601.6239 Ops/s $\color{#d91a1a}-2.83\%$
test_ddpg_speed 14.3400ms 3.1210ms 320.4066 Ops/s 329.3058 Ops/s $\color{#d91a1a}-2.70\%$
test_sac_speed 19.4141ms 8.8588ms 112.8820 Ops/s 117.3743 Ops/s $\color{#d91a1a}-3.83\%$
test_redq_speed 25.5976ms 16.9885ms 58.8633 Ops/s 59.8819 Ops/s $\color{#d91a1a}-1.70\%$
test_redq_deprec_speed 23.6720ms 14.8932ms 67.1448 Ops/s 66.8866 Ops/s $\color{#35bf28}+0.39\%$
test_td3_speed 17.6853ms 9.0354ms 110.6753 Ops/s 115.0878 Ops/s $\color{#d91a1a}-3.83\%$
test_cql_speed 36.0697ms 35.5006ms 28.1685 Ops/s 26.9149 Ops/s $\color{#35bf28}+4.66\%$
test_a2c_speed 16.1904ms 7.8283ms 127.7414 Ops/s 122.8923 Ops/s $\color{#35bf28}+3.95\%$
test_ppo_speed 16.8202ms 8.0889ms 123.6269 Ops/s 117.2104 Ops/s $\textbf{\color{#35bf28}+5.47\%}$
test_reinforce_speed 16.1208ms 7.4632ms 133.9909 Ops/s 139.3825 Ops/s $\color{#d91a1a}-3.87\%$
test_iql_speed 44.2583ms 32.4250ms 30.8404 Ops/s 29.9284 Ops/s $\color{#35bf28}+3.05\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.4883ms 1.9831ms 504.2561 Ops/s 504.7505 Ops/s $\color{#d91a1a}-0.10\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 4.1343ms 2.1345ms 468.4952 Ops/s 487.6802 Ops/s $\color{#d91a1a}-3.93\%$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 3.1440ms 2.1454ms 466.1068 Ops/s 475.8984 Ops/s $\color{#d91a1a}-2.06\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.7766ms 1.9740ms 506.5899 Ops/s 472.1125 Ops/s $\textbf{\color{#35bf28}+7.30\%}$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 3.4051ms 1.9909ms 502.2819 Ops/s 464.6833 Ops/s $\textbf{\color{#35bf28}+8.09\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 2.9709ms 1.9819ms 504.5655 Ops/s 464.5699 Ops/s $\textbf{\color{#35bf28}+8.61\%}$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.4603ms 1.8734ms 533.7868 Ops/s 513.5727 Ops/s $\color{#35bf28}+3.94\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 6.3552ms 2.0854ms 479.5165 Ops/s 482.6055 Ops/s $\color{#d91a1a}-0.64\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 3.7075ms 2.1618ms 462.5836 Ops/s 476.1613 Ops/s $\color{#d91a1a}-2.85\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.2114ms 1.9486ms 513.1911 Ops/s 510.8629 Ops/s $\color{#35bf28}+0.46\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.9165ms 2.1683ms 461.1998 Ops/s 475.5778 Ops/s $\color{#d91a1a}-3.02\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.3967ms 2.1804ms 458.6231 Ops/s 472.6347 Ops/s $\color{#d91a1a}-2.96\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.6360ms 1.9320ms 517.5989 Ops/s 502.7030 Ops/s $\color{#35bf28}+2.96\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.8519ms 2.0377ms 490.7447 Ops/s 477.8606 Ops/s $\color{#35bf28}+2.70\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 3.4910ms 2.1947ms 455.6492 Ops/s 481.3078 Ops/s $\textbf{\color{#d91a1a}-5.33\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.5872ms 2.0374ms 490.8157 Ops/s 410.1115 Ops/s $\textbf{\color{#35bf28}+19.68\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 4.2693ms 2.2249ms 449.4623 Ops/s 477.1261 Ops/s $\textbf{\color{#d91a1a}-5.80\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.1357s 2.4891ms 401.7585 Ops/s 470.1069 Ops/s $\textbf{\color{#d91a1a}-14.54\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.2875s 23.6948ms 42.2033 Ops/s 40.0133 Ops/s $\textbf{\color{#35bf28}+5.47\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 0.1361s 22.9739ms 43.5277 Ops/s 48.5420 Ops/s $\textbf{\color{#d91a1a}-10.33\%}$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.1349s 20.9128ms 47.8176 Ops/s 43.6424 Ops/s $\textbf{\color{#35bf28}+9.57\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1336s 23.2362ms 43.0363 Ops/s 43.2932 Ops/s $\color{#d91a1a}-0.59\%$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1401s 20.9246ms 47.7907 Ops/s 48.5107 Ops/s $\color{#d91a1a}-1.48\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.1365s 23.1030ms 43.2843 Ops/s 43.2616 Ops/s $\color{#35bf28}+0.05\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1430s 20.6665ms 48.3876 Ops/s 48.5860 Ops/s $\color{#d91a1a}-0.41\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.1284s 22.1677ms 45.1107 Ops/s 42.5552 Ops/s $\textbf{\color{#35bf28}+6.01\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 0.1326s 20.3566ms 49.1240 Ops/s 47.5785 Ops/s $\color{#35bf28}+3.25\%$

@github-actions GitHub Actions
Copy link

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 60.5903ms 58.1413ms 17.1995 Ops/s 16.3483 Ops/s $\textbf{\color{#35bf28}+5.21\%}$
test_sync 36.8440ms 32.0837ms 31.1685 Ops/s 30.3216 Ops/s $\color{#35bf28}+2.79\%$
test_async 61.1336ms 28.6779ms 34.8701 Ops/s 34.2917 Ops/s $\color{#35bf28}+1.69\%$
test_simple 0.4702s 0.4151s 2.4090 Ops/s 2.3744 Ops/s $\color{#35bf28}+1.46\%$
test_transformed 0.6160s 0.5767s 1.7339 Ops/s 1.7231 Ops/s $\color{#35bf28}+0.62\%$
test_serial 1.2268s 1.1827s 0.8455 Ops/s 0.8247 Ops/s $\color{#35bf28}+2.53\%$
test_parallel 1.1223s 1.0769s 0.9286 Ops/s 0.9075 Ops/s $\color{#35bf28}+2.31\%$
test_step_mdp_speed[True-True-True-True-True] 0.1593ms 20.1659μs 49.5888 KOps/s 49.7495 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[True-True-True-True-False] 46.5660μs 11.9136μs 83.9378 KOps/s 84.2361 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-True-True-False-True] 40.6060μs 12.2638μs 81.5409 KOps/s 81.6253 KOps/s $\color{#d91a1a}-0.10\%$
test_step_mdp_speed[True-True-True-False-False] 53.1390μs 7.2704μs 137.5437 KOps/s 138.9167 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[True-True-False-True-True] 55.0930μs 21.2490μs 47.0611 KOps/s 46.8548 KOps/s $\color{#35bf28}+0.44\%$
test_step_mdp_speed[True-True-False-True-False] 43.5510μs 13.3076μs 75.1451 KOps/s 75.8050 KOps/s $\color{#d91a1a}-0.87\%$
test_step_mdp_speed[True-True-False-False-True] 37.9210μs 13.4998μs 74.0751 KOps/s 74.4438 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-True-False-False-False] 27.4510μs 8.3824μs 119.2980 KOps/s 119.2872 KOps/s $+0.01\%$
test_step_mdp_speed[True-False-True-True-True] 46.4760μs 22.4617μs 44.5201 KOps/s 44.4428 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-False-True-True-False] 37.4100μs 14.3295μs 69.7861 KOps/s 69.4476 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[True-False-True-False-True] 46.2260μs 13.4988μs 74.0806 KOps/s 74.7146 KOps/s $\color{#d91a1a}-0.85\%$
test_step_mdp_speed[True-False-True-False-False] 34.3940μs 8.3748μs 119.4060 KOps/s 120.1580 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[True-False-False-True-True] 45.9860μs 23.6181μs 42.3405 KOps/s 42.4045 KOps/s $\color{#d91a1a}-0.15\%$
test_step_mdp_speed[True-False-False-True-False] 47.0680μs 15.4548μs 64.7047 KOps/s 65.4626 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[True-False-False-False-True] 43.4910μs 14.6656μs 68.1868 KOps/s 69.4174 KOps/s $\color{#d91a1a}-1.77\%$
test_step_mdp_speed[True-False-False-False-False] 31.2380μs 9.4770μs 105.5185 KOps/s 105.2702 KOps/s $\color{#35bf28}+0.24\%$
test_step_mdp_speed[False-True-True-True-True] 48.2100μs 22.4413μs 44.5606 KOps/s 44.6273 KOps/s $\color{#d91a1a}-0.15\%$
test_step_mdp_speed[False-True-True-True-False] 37.0090μs 14.4215μs 69.3408 KOps/s 68.7526 KOps/s $\color{#35bf28}+0.86\%$
test_step_mdp_speed[False-True-True-False-True] 43.9820μs 15.3502μs 65.1459 KOps/s 64.2903 KOps/s $\color{#35bf28}+1.33\%$
test_step_mdp_speed[False-True-True-False-False] 31.4280μs 9.4201μs 106.1555 KOps/s 104.6490 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[False-True-False-True-True] 56.4850μs 23.4341μs 42.6728 KOps/s 42.3395 KOps/s $\color{#35bf28}+0.79\%$
test_step_mdp_speed[False-True-False-True-False] 37.4900μs 15.3812μs 65.0146 KOps/s 65.5677 KOps/s $\color{#d91a1a}-0.84\%$
test_step_mdp_speed[False-True-False-False-True] 46.1050μs 16.3900μs 61.0128 KOps/s 60.2662 KOps/s $\color{#35bf28}+1.24\%$
test_step_mdp_speed[False-True-False-False-False] 32.3300μs 10.5598μs 94.6990 KOps/s 96.4313 KOps/s $\color{#d91a1a}-1.80\%$
test_step_mdp_speed[False-False-True-True-True] 49.9630μs 24.6829μs 40.5139 KOps/s 40.0597 KOps/s $\color{#35bf28}+1.13\%$
test_step_mdp_speed[False-False-True-True-False] 49.3720μs 16.6616μs 60.0184 KOps/s 61.1871 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[False-False-True-False-True] 55.3630μs 16.4579μs 60.7611 KOps/s 60.4288 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-False-True-False-False] 37.3190μs 10.5211μs 95.0471 KOps/s 96.0564 KOps/s $\color{#d91a1a}-1.05\%$
test_step_mdp_speed[False-False-False-True-True] 52.6980μs 25.7328μs 38.8609 KOps/s 39.0325 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[False-False-False-True-False] 64.3590μs 17.4471μs 57.3162 KOps/s 57.4855 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[False-False-False-False-True] 54.7010μs 17.2857μs 57.8514 KOps/s 57.6930 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[False-False-False-False-False] 40.7260μs 11.5054μs 86.9161 KOps/s 87.8380 KOps/s $\color{#d91a1a}-1.05\%$
test_values[generalized_advantage_estimate-True-True] 14.5007ms 11.7581ms 85.0477 Ops/s 84.6312 Ops/s $\color{#35bf28}+0.49\%$
test_values[vec_generalized_advantage_estimate-True-True] 68.7701ms 28.9808ms 34.5056 Ops/s 36.7599 Ops/s $\textbf{\color{#d91a1a}-6.13\%}$
test_values[td0_return_estimate-False-False] 0.2582ms 0.1764ms 5.6683 KOps/s 5.4517 KOps/s $\color{#35bf28}+3.97\%$
test_values[td1_return_estimate-False-False] 25.6972ms 24.6192ms 40.6188 Ops/s 40.0015 Ops/s $\color{#35bf28}+1.54\%$
test_values[vec_td1_return_estimate-False-False] 37.0752ms 27.2127ms 36.7476 Ops/s 37.3256 Ops/s $\color{#d91a1a}-1.55\%$
test_values[td_lambda_return_estimate-True-False] 35.3229ms 34.2485ms 29.1983 Ops/s 28.3906 Ops/s $\color{#35bf28}+2.84\%$
test_values[vec_td_lambda_return_estimate-True-False] 30.3714ms 27.0176ms 37.0129 Ops/s 36.9996 Ops/s $\color{#35bf28}+0.04\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.5349ms 7.7259ms 129.4344 Ops/s 127.8036 Ops/s $\color{#35bf28}+1.28\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.0406ms 1.8139ms 551.2994 Ops/s 490.7431 Ops/s $\textbf{\color{#35bf28}+12.34\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 11.6288ms 0.4395ms 2.2755 KOps/s 2.2959 KOps/s $\color{#d91a1a}-0.89\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 46.9200ms 40.2511ms 24.8440 Ops/s 25.0286 Ops/s $\color{#d91a1a}-0.74\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 13.5452ms 2.5227ms 396.3999 Ops/s 392.0811 Ops/s $\color{#35bf28}+1.10\%$
test_dqn_speed 9.7170ms 1.5998ms 625.0632 Ops/s 627.3374 Ops/s $\color{#d91a1a}-0.36\%$
test_ddpg_speed 14.7019ms 2.7777ms 360.0087 Ops/s 338.7013 Ops/s $\textbf{\color{#35bf28}+6.29\%}$
test_sac_speed 19.0858ms 8.1523ms 122.6653 Ops/s 119.1859 Ops/s $\color{#35bf28}+2.92\%$
test_redq_speed 25.7390ms 15.2165ms 65.7181 Ops/s 64.2037 Ops/s $\color{#35bf28}+2.36\%$
test_redq_deprec_speed 24.8920ms 13.7014ms 72.9854 Ops/s 72.4382 Ops/s $\color{#35bf28}+0.76\%$
test_td3_speed 9.4759ms 8.3626ms 119.5794 Ops/s 117.7283 Ops/s $\color{#35bf28}+1.57\%$
test_cql_speed 46.0299ms 35.3386ms 28.2977 Ops/s 27.7500 Ops/s $\color{#35bf28}+1.97\%$
test_a2c_speed 18.6569ms 7.6871ms 130.0882 Ops/s 120.5743 Ops/s $\textbf{\color{#35bf28}+7.89\%}$
test_ppo_speed 19.5811ms 8.0993ms 123.4679 Ops/s 116.5656 Ops/s $\textbf{\color{#35bf28}+5.92\%}$
test_reinforce_speed 18.1293ms 6.8679ms 145.6044 Ops/s 146.1022 Ops/s $\color{#d91a1a}-0.34\%$
test_iql_speed 42.0124ms 31.7161ms 31.5297 Ops/s 31.0610 Ops/s $\color{#35bf28}+1.51\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 1.9247ms 1.5903ms 628.8203 Ops/s 612.7152 Ops/s $\color{#35bf28}+2.63\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.7395ms 1.6562ms 603.7783 Ops/s 566.6092 Ops/s $\textbf{\color{#35bf28}+6.56\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.1229s 1.9064ms 524.5404 Ops/s 561.2998 Ops/s $\textbf{\color{#d91a1a}-6.55\%}$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.0859ms 1.5618ms 640.2758 Ops/s 493.3680 Ops/s $\textbf{\color{#35bf28}+29.78\%}$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.3349ms 1.6954ms 589.8194 Ops/s 564.1752 Ops/s $\color{#35bf28}+4.55\%$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 2.5951ms 1.6915ms 591.1959 Ops/s 565.2856 Ops/s $\color{#35bf28}+4.58\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.1667ms 1.5677ms 637.8717 Ops/s 611.2114 Ops/s $\color{#35bf28}+4.36\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.6328ms 1.6602ms 602.3490 Ops/s 578.9001 Ops/s $\color{#35bf28}+4.05\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 3.4788ms 1.7748ms 563.4293 Ops/s 567.7642 Ops/s $\color{#d91a1a}-0.76\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 1.8870ms 1.6653ms 600.5064 Ops/s 616.7210 Ops/s $\color{#d91a1a}-2.63\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.6075ms 1.7546ms 569.9417 Ops/s 563.8379 Ops/s $\color{#35bf28}+1.08\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.5146ms 1.7829ms 560.8739 Ops/s 562.3169 Ops/s $\color{#d91a1a}-0.26\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.1978ms 1.6074ms 622.1190 Ops/s 606.7001 Ops/s $\color{#35bf28}+2.54\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 3.3150ms 1.7010ms 587.8955 Ops/s 576.9010 Ops/s $\color{#35bf28}+1.91\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 3.3381ms 1.6969ms 589.3047 Ops/s 558.6875 Ops/s $\textbf{\color{#35bf28}+5.48\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.0700ms 1.5587ms 641.5490 Ops/s 609.0010 Ops/s $\textbf{\color{#35bf28}+5.34\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.5476ms 1.7043ms 586.7553 Ops/s 564.3367 Ops/s $\color{#35bf28}+3.97\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 2.6457ms 1.6960ms 589.6288 Ops/s 561.1075 Ops/s $\textbf{\color{#35bf28}+5.08\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.2018s 19.5610ms 51.1220 Ops/s 44.7078 Ops/s $\textbf{\color{#35bf28}+14.35\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 0.1071s 18.9940ms 52.6483 Ops/s 54.9300 Ops/s $\color{#d91a1a}-4.15\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.1130s 19.5656ms 51.1102 Ops/s 50.2652 Ops/s $\color{#35bf28}+1.68\%$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1065s 15.3528ms 65.1348 Ops/s 55.2407 Ops/s $\textbf{\color{#35bf28}+17.91\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1032s 18.7823ms 53.2416 Ops/s 49.0782 Ops/s $\textbf{\color{#35bf28}+8.48\%}$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.1021s 16.7154ms 59.8252 Ops/s 48.3588 Ops/s $\textbf{\color{#35bf28}+23.71\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1188s 19.5394ms 51.1787 Ops/s 55.0666 Ops/s $\textbf{\color{#d91a1a}-7.06\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.1054s 17.3381ms 57.6766 Ops/s 48.9267 Ops/s $\textbf{\color{#35bf28}+17.88\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 0.1081s 18.9637ms 52.7323 Ops/s 55.6592 Ops/s $\textbf{\color{#d91a1a}-5.26\%}$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Milestone
No milestone
Development

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