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

[BugFix, Test] Fix torch.vmap call in RNN tests #1749

Merged
merged 1 commit into from
Dec 15, 2023
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Dec 15, 2023

No description provided.

Copy link

pytorch-bot bot commented Dec 15, 2023

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1749

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

✅ You can merge normally! (27 Unrelated Failures)

As of commit 960f5d1 with merge base b3d2aa6 (image):

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.

@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 Dec 15, 2023
@vmoens vmoens marked this pull request as ready for review December 15, 2023 12:16
@vmoens vmoens merged commit 08f0bed into main Dec 15, 2023
@vmoens vmoens deleted the fix-vmap-lstm-test branch December 15, 2023 12:16
@vmoens vmoens added the Tests Incomplete or broken unit tests label Dec 15, 2023
Copy link

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}5$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 62.8076ms 62.4393ms 16.0155 Ops/s 15.1526 Ops/s $\textbf{\color{#35bf28}+5.70\%}$
test_sync 47.8621ms 40.0243ms 24.9848 Ops/s 28.2639 Ops/s $\textbf{\color{#d91a1a}-11.60\%}$
test_async 66.2293ms 33.5234ms 29.8299 Ops/s 30.4774 Ops/s $\color{#d91a1a}-2.12\%$
test_simple 0.4996s 0.4468s 2.2384 Ops/s 2.3024 Ops/s $\color{#d91a1a}-2.78\%$
test_transformed 0.6625s 0.6163s 1.6225 Ops/s 1.6763 Ops/s $\color{#d91a1a}-3.21\%$
test_serial 1.4350s 1.3858s 0.7216 Ops/s 0.7381 Ops/s $\color{#d91a1a}-2.24\%$
test_parallel 1.3839s 1.3254s 0.7545 Ops/s 0.7574 Ops/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[True-True-True-True-True] 0.2561ms 21.4897μs 46.5339 KOps/s 45.9318 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[True-True-True-True-False] 48.7220μs 12.9777μs 77.0551 KOps/s 75.3643 KOps/s $\color{#35bf28}+2.24\%$
test_step_mdp_speed[True-True-True-False-True] 33.6030μs 12.8733μs 77.6799 KOps/s 77.5502 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-True-True-False-False] 31.6690μs 7.7691μs 128.7143 KOps/s 128.5190 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[True-True-False-True-True] 67.4160μs 22.9657μs 43.5432 KOps/s 43.2449 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[True-True-False-True-False] 50.6650μs 14.3357μs 69.7559 KOps/s 69.0043 KOps/s $\color{#35bf28}+1.09\%$
test_step_mdp_speed[True-True-False-False-True] 37.0700μs 13.9750μs 71.5564 KOps/s 70.8159 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-True-False-False-False] 34.6150μs 8.9624μs 111.5767 KOps/s 111.2325 KOps/s $\color{#35bf28}+0.31\%$
test_step_mdp_speed[True-False-True-True-True] 68.7880μs 24.3442μs 41.0776 KOps/s 41.1792 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[True-False-True-True-False] 68.5080μs 15.6103μs 64.0604 KOps/s 63.2762 KOps/s $\color{#35bf28}+1.24\%$
test_step_mdp_speed[True-False-True-False-True] 61.4350μs 14.6749μs 68.1438 KOps/s 71.4461 KOps/s $\color{#d91a1a}-4.62\%$
test_step_mdp_speed[True-False-True-False-False] 36.5980μs 8.8602μs 112.8639 KOps/s 111.0069 KOps/s $\color{#35bf28}+1.67\%$
test_step_mdp_speed[True-False-False-True-True] 72.4460μs 25.2419μs 39.6166 KOps/s 39.0399 KOps/s $\color{#35bf28}+1.48\%$
test_step_mdp_speed[True-False-False-True-False] 42.4190μs 16.6475μs 60.0691 KOps/s 58.2320 KOps/s $\color{#35bf28}+3.15\%$
test_step_mdp_speed[True-False-False-False-True] 49.4220μs 15.3227μs 65.2625 KOps/s 66.2650 KOps/s $\color{#d91a1a}-1.51\%$
test_step_mdp_speed[True-False-False-False-False] 39.5340μs 10.0545μs 99.4576 KOps/s 98.1677 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[False-True-True-True-True] 50.7550μs 24.2190μs 41.2898 KOps/s 41.2527 KOps/s $\color{#35bf28}+0.09\%$
test_step_mdp_speed[False-True-True-True-False] 47.5090μs 15.7782μs 63.3786 KOps/s 63.2676 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-True-True-False-True] 45.2350μs 16.2675μs 61.4722 KOps/s 61.4435 KOps/s $\color{#35bf28}+0.05\%$
test_step_mdp_speed[False-True-True-False-False] 36.2880μs 10.1836μs 98.1969 KOps/s 97.9691 KOps/s $\color{#35bf28}+0.23\%$
test_step_mdp_speed[False-True-False-True-True] 73.9880μs 25.3311μs 39.4771 KOps/s 39.2229 KOps/s $\color{#35bf28}+0.65\%$
test_step_mdp_speed[False-True-False-True-False] 46.1670μs 16.8802μs 59.2410 KOps/s 58.6539 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[False-True-False-False-True] 44.4530μs 17.4252μs 57.3883 KOps/s 56.6256 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[False-True-False-False-False] 35.9870μs 11.3162μs 88.3686 KOps/s 87.5879 KOps/s $\color{#35bf28}+0.89\%$
test_step_mdp_speed[False-False-True-True-True] 75.1910μs 26.7101μs 37.4391 KOps/s 37.0483 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[False-False-True-True-False] 47.3690μs 18.3027μs 54.6369 KOps/s 54.0563 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-False-True-False-True] 45.0940μs 17.3046μs 57.7881 KOps/s 57.0386 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[False-False-True-False-False] 60.2520μs 11.4440μs 87.3822 KOps/s 86.6602 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-False-False-True-True] 65.9630μs 27.5827μs 36.2546 KOps/s 35.7494 KOps/s $\color{#35bf28}+1.41\%$
test_step_mdp_speed[False-False-False-True-False] 71.4530μs 19.4284μs 51.4709 KOps/s 51.6640 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[False-False-False-False-True] 63.2980μs 18.1446μs 55.1128 KOps/s 53.6618 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[False-False-False-False-False] 43.4610μs 12.3743μs 80.8125 KOps/s 80.3298 KOps/s $\color{#35bf28}+0.60\%$
test_values[generalized_advantage_estimate-True-True] 15.5545ms 12.0308ms 83.1197 Ops/s 82.0772 Ops/s $\color{#35bf28}+1.27\%$
test_values[vec_generalized_advantage_estimate-True-True] 29.3489ms 27.5844ms 36.2523 Ops/s 37.3307 Ops/s $\color{#d91a1a}-2.89\%$
test_values[td0_return_estimate-False-False] 0.2574ms 0.1762ms 5.6756 KOps/s 5.6760 KOps/s $-0.01\%$
test_values[td1_return_estimate-False-False] 27.0316ms 25.8990ms 38.6116 Ops/s 38.8456 Ops/s $\color{#d91a1a}-0.60\%$
test_values[vec_td1_return_estimate-False-False] 38.9585ms 27.8653ms 35.8869 Ops/s 36.7606 Ops/s $\color{#d91a1a}-2.38\%$
test_values[td_lambda_return_estimate-True-False] 40.6832ms 36.3487ms 27.5113 Ops/s 28.0228 Ops/s $\color{#d91a1a}-1.83\%$
test_values[vec_td_lambda_return_estimate-True-False] 37.4609ms 27.8805ms 35.8673 Ops/s 36.1307 Ops/s $\color{#d91a1a}-0.73\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.1490ms 8.1151ms 123.2276 Ops/s 118.1126 Ops/s $\color{#35bf28}+4.33\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 4.4376ms 1.7421ms 574.0098 Ops/s 547.9495 Ops/s $\color{#35bf28}+4.76\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 17.1286ms 0.4494ms 2.2249 KOps/s 2.2428 KOps/s $\color{#d91a1a}-0.80\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 97.5582ms 43.5046ms 22.9861 Ops/s 25.0742 Ops/s $\textbf{\color{#d91a1a}-8.33\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 13.1552ms 2.6493ms 377.4518 Ops/s 375.1538 Ops/s $\color{#35bf28}+0.61\%$
test_dqn_speed 12.7713ms 1.6453ms 607.7835 Ops/s 631.6965 Ops/s $\color{#d91a1a}-3.79\%$
test_ddpg_speed 14.1834ms 3.6481ms 274.1134 Ops/s 279.3947 Ops/s $\color{#d91a1a}-1.89\%$
test_sac_speed 20.8433ms 10.2926ms 97.1574 Ops/s 98.3404 Ops/s $\color{#d91a1a}-1.20\%$
test_redq_speed 30.8718ms 19.2190ms 52.0318 Ops/s 52.9653 Ops/s $\color{#d91a1a}-1.76\%$
test_redq_deprec_speed 26.4381ms 15.1365ms 66.0656 Ops/s 66.7137 Ops/s $\color{#d91a1a}-0.97\%$
test_td3_speed 17.7734ms 10.5285ms 94.9799 Ops/s 94.9441 Ops/s $\color{#35bf28}+0.04\%$
test_cql_speed 43.1089ms 41.0103ms 24.3841 Ops/s 24.2604 Ops/s $\color{#35bf28}+0.51\%$
test_a2c_speed 20.6110ms 8.6492ms 115.6177 Ops/s 115.4410 Ops/s $\color{#35bf28}+0.15\%$
test_ppo_speed 20.5676ms 8.9425ms 111.8256 Ops/s 111.8898 Ops/s $\color{#d91a1a}-0.06\%$
test_reinforce_speed 19.1376ms 7.7254ms 129.4437 Ops/s 129.7481 Ops/s $\color{#d91a1a}-0.23\%$
test_iql_speed 47.0078ms 35.8034ms 27.9303 Ops/s 26.6412 Ops/s $\color{#35bf28}+4.84\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.8089ms 1.8494ms 540.7050 Ops/s 536.2294 Ops/s $\color{#35bf28}+0.83\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 97.1502ms 2.1169ms 472.3926 Ops/s 523.8074 Ops/s $\textbf{\color{#d91a1a}-9.82\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 2.7381ms 1.9215ms 520.4337 Ops/s 509.0000 Ops/s $\color{#35bf28}+2.25\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.4165ms 1.8443ms 542.1990 Ops/s 548.8476 Ops/s $\color{#d91a1a}-1.21\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.1100s 2.1481ms 465.5305 Ops/s 525.4201 Ops/s $\textbf{\color{#d91a1a}-11.40\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 2.6940ms 1.9223ms 520.2203 Ops/s 517.4729 Ops/s $\color{#35bf28}+0.53\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.9204ms 1.8580ms 538.2257 Ops/s 550.8556 Ops/s $\color{#d91a1a}-2.29\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.1006s 2.1280ms 469.9359 Ops/s 520.9244 Ops/s $\textbf{\color{#d91a1a}-9.79\%}$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 2.6111ms 1.9229ms 520.0508 Ops/s 516.6992 Ops/s $\color{#35bf28}+0.65\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.9200ms 1.8647ms 536.2792 Ops/s 545.1883 Ops/s $\color{#d91a1a}-1.63\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.1015s 2.1184ms 472.0507 Ops/s 516.6658 Ops/s $\textbf{\color{#d91a1a}-8.64\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 3.1451ms 1.9252ms 519.4134 Ops/s 524.6772 Ops/s $\color{#d91a1a}-1.00\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.2732ms 1.8586ms 538.0489 Ops/s 549.9556 Ops/s $\color{#d91a1a}-2.17\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 98.7475ms 2.1240ms 470.8172 Ops/s 517.8122 Ops/s $\textbf{\color{#d91a1a}-9.08\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 2.6841ms 1.9207ms 520.6324 Ops/s 459.9442 Ops/s $\textbf{\color{#35bf28}+13.19\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.9523ms 1.8604ms 537.5328 Ops/s 546.7617 Ops/s $\color{#d91a1a}-1.69\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.1018s 2.1347ms 468.4467 Ops/s 524.6523 Ops/s $\textbf{\color{#d91a1a}-10.71\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 3.0823ms 1.9389ms 515.7632 Ops/s 473.8928 Ops/s $\textbf{\color{#35bf28}+8.84\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1430s 16.6582ms 60.0303 Ops/s 65.0311 Ops/s $\textbf{\color{#d91a1a}-7.69\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 99.1697ms 14.0789ms 71.0284 Ops/s 64.1647 Ops/s $\textbf{\color{#35bf28}+10.70\%}$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.1007s 15.9766ms 62.5916 Ops/s 72.3814 Ops/s $\textbf{\color{#d91a1a}-13.53\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 98.6974ms 15.8126ms 63.2405 Ops/s 57.7917 Ops/s $\textbf{\color{#35bf28}+9.43\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 98.1289ms 15.7747ms 63.3927 Ops/s 63.7324 Ops/s $\color{#d91a1a}-0.53\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 97.3867ms 15.7415ms 63.5263 Ops/s 62.9115 Ops/s $\color{#35bf28}+0.98\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 97.8906ms 15.6483ms 63.9047 Ops/s 64.2612 Ops/s $\color{#d91a1a}-0.55\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 98.0763ms 15.7963ms 63.3059 Ops/s 64.2036 Ops/s $\color{#d91a1a}-1.40\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 97.7369ms 15.5310ms 64.3875 Ops/s 63.4531 Ops/s $\color{#35bf28}+1.47\%$

Copy link

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

Total Benchmarks: 92. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1225s 0.1220s 8.1972 Ops/s 8.1116 Ops/s $\color{#35bf28}+1.06\%$
test_sync 0.1052s 0.1022s 9.7802 Ops/s 9.7395 Ops/s $\color{#35bf28}+0.42\%$
test_async 0.2780s 99.3409ms 10.0663 Ops/s 9.9743 Ops/s $\color{#35bf28}+0.92\%$
test_single_pixels 0.1315s 0.1312s 7.6192 Ops/s 6.8149 Ops/s $\textbf{\color{#35bf28}+11.80\%}$
test_sync_pixels 97.7161ms 95.9467ms 10.4224 Ops/s 10.3017 Ops/s $\color{#35bf28}+1.17\%$
test_async_pixels 0.2440s 90.2837ms 11.0762 Ops/s 10.9162 Ops/s $\color{#35bf28}+1.47\%$
test_simple 0.9557s 0.8882s 1.1259 Ops/s 1.1056 Ops/s $\color{#35bf28}+1.83\%$
test_transformed 1.2010s 1.1367s 0.8798 Ops/s 0.8842 Ops/s $\color{#d91a1a}-0.50\%$
test_serial 2.5702s 2.5081s 0.3987 Ops/s 0.3947 Ops/s $\color{#35bf28}+1.02\%$
test_parallel 2.5688s 2.4917s 0.4013 Ops/s 0.4002 Ops/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[True-True-True-True-True] 0.1100ms 32.0479μs 31.2033 KOps/s 30.3586 KOps/s $\color{#35bf28}+2.78\%$
test_step_mdp_speed[True-True-True-True-False] 42.1710μs 19.0853μs 52.3962 KOps/s 50.7519 KOps/s $\color{#35bf28}+3.24\%$
test_step_mdp_speed[True-True-True-False-True] 44.0600μs 18.1625μs 55.0585 KOps/s 52.6591 KOps/s $\color{#35bf28}+4.56\%$
test_step_mdp_speed[True-True-True-False-False] 31.0500μs 10.8837μs 91.8806 KOps/s 87.2812 KOps/s $\textbf{\color{#35bf28}+5.27\%}$
test_step_mdp_speed[True-True-False-True-True] 62.7020μs 33.9303μs 29.4722 KOps/s 28.5018 KOps/s $\color{#35bf28}+3.40\%$
test_step_mdp_speed[True-True-False-True-False] 45.8100μs 20.6211μs 48.4940 KOps/s 46.4573 KOps/s $\color{#35bf28}+4.38\%$
test_step_mdp_speed[True-True-False-False-True] 45.2910μs 20.1269μs 49.6847 KOps/s 48.2051 KOps/s $\color{#35bf28}+3.07\%$
test_step_mdp_speed[True-True-False-False-False] 54.0310μs 12.8257μs 77.9683 KOps/s 76.0390 KOps/s $\color{#35bf28}+2.54\%$
test_step_mdp_speed[True-False-True-True-True] 65.6110μs 35.8300μs 27.9096 KOps/s 26.8490 KOps/s $\color{#35bf28}+3.95\%$
test_step_mdp_speed[True-False-True-True-False] 48.0900μs 22.6789μs 44.0938 KOps/s 41.9339 KOps/s $\textbf{\color{#35bf28}+5.15\%}$
test_step_mdp_speed[True-False-True-False-True] 46.8110μs 20.0614μs 49.8469 KOps/s 48.3588 KOps/s $\color{#35bf28}+3.08\%$
test_step_mdp_speed[True-False-True-False-False] 79.5610μs 12.7517μs 78.4211 KOps/s 75.2299 KOps/s $\color{#35bf28}+4.24\%$
test_step_mdp_speed[True-False-False-True-True] 66.0610μs 36.9649μs 27.0527 KOps/s 25.5737 KOps/s $\textbf{\color{#35bf28}+5.78\%}$
test_step_mdp_speed[True-False-False-True-False] 50.0300μs 24.4241μs 40.9432 KOps/s 39.1238 KOps/s $\color{#35bf28}+4.65\%$
test_step_mdp_speed[True-False-False-False-True] 59.0810μs 21.7028μs 46.0769 KOps/s 44.1386 KOps/s $\color{#35bf28}+4.39\%$
test_step_mdp_speed[True-False-False-False-False] 34.8190μs 14.3636μs 69.6204 KOps/s 64.9933 KOps/s $\textbf{\color{#35bf28}+7.12\%}$
test_step_mdp_speed[False-True-True-True-True] 0.1025ms 35.1501μs 28.4494 KOps/s 27.1386 KOps/s $\color{#35bf28}+4.83\%$
test_step_mdp_speed[False-True-True-True-False] 43.0500μs 22.5369μs 44.3716 KOps/s 41.7999 KOps/s $\textbf{\color{#35bf28}+6.15\%}$
test_step_mdp_speed[False-True-True-False-True] 60.6900μs 24.0510μs 41.5783 KOps/s 39.3361 KOps/s $\textbf{\color{#35bf28}+5.70\%}$
test_step_mdp_speed[False-True-True-False-False] 38.6710μs 14.6533μs 68.2441 KOps/s 63.6217 KOps/s $\textbf{\color{#35bf28}+7.27\%}$
test_step_mdp_speed[False-True-False-True-True] 0.1064ms 38.0699μs 26.2675 KOps/s 25.2716 KOps/s $\color{#35bf28}+3.94\%$
test_step_mdp_speed[False-True-False-True-False] 46.2910μs 24.3170μs 41.1234 KOps/s 38.6854 KOps/s $\textbf{\color{#35bf28}+6.30\%}$
test_step_mdp_speed[False-True-False-False-True] 95.3420μs 25.7404μs 38.8494 KOps/s 36.8558 KOps/s $\textbf{\color{#35bf28}+5.41\%}$
test_step_mdp_speed[False-True-False-False-False] 40.8110μs 16.1846μs 61.7873 KOps/s 57.7907 KOps/s $\textbf{\color{#35bf28}+6.92\%}$
test_step_mdp_speed[False-False-True-True-True] 69.3010μs 39.1597μs 25.5365 KOps/s 24.4304 KOps/s $\color{#35bf28}+4.53\%$
test_step_mdp_speed[False-False-True-True-False] 51.4110μs 26.2563μs 38.0862 KOps/s 36.2705 KOps/s $\textbf{\color{#35bf28}+5.01\%}$
test_step_mdp_speed[False-False-True-False-True] 50.9500μs 26.2068μs 38.1581 KOps/s 36.9246 KOps/s $\color{#35bf28}+3.34\%$
test_step_mdp_speed[False-False-True-False-False] 77.6010μs 16.2805μs 61.4230 KOps/s 56.2097 KOps/s $\textbf{\color{#35bf28}+9.27\%}$
test_step_mdp_speed[False-False-False-True-True] 72.0510μs 40.9749μs 24.4052 KOps/s 23.5806 KOps/s $\color{#35bf28}+3.50\%$
test_step_mdp_speed[False-False-False-True-False] 53.6110μs 28.0367μs 35.6675 KOps/s 33.8049 KOps/s $\textbf{\color{#35bf28}+5.51\%}$
test_step_mdp_speed[False-False-False-False-True] 50.0500μs 27.2383μs 36.7130 KOps/s 34.9153 KOps/s $\textbf{\color{#35bf28}+5.15\%}$
test_step_mdp_speed[False-False-False-False-False] 39.5200μs 18.0202μs 55.4931 KOps/s 51.7892 KOps/s $\textbf{\color{#35bf28}+7.15\%}$
test_values[generalized_advantage_estimate-True-True] 26.4578ms 25.7166ms 38.8854 Ops/s 39.2919 Ops/s $\color{#d91a1a}-1.03\%$
test_values[vec_generalized_advantage_estimate-True-True] 83.1391ms 3.2432ms 308.3330 Ops/s 303.4988 Ops/s $\color{#35bf28}+1.59\%$
test_values[td0_return_estimate-False-False] 0.1001ms 64.6685μs 15.4635 KOps/s 15.2039 KOps/s $\color{#35bf28}+1.71\%$
test_values[td1_return_estimate-False-False] 55.1457ms 54.6413ms 18.3012 Ops/s 17.6576 Ops/s $\color{#35bf28}+3.64\%$
test_values[vec_td1_return_estimate-False-False] 2.1025ms 1.7825ms 560.9994 Ops/s 560.3275 Ops/s $\color{#35bf28}+0.12\%$
test_values[td_lambda_return_estimate-True-False] 92.5892ms 87.6684ms 11.4066 Ops/s 11.4279 Ops/s $\color{#d91a1a}-0.19\%$
test_values[vec_td_lambda_return_estimate-True-False] 2.1112ms 1.7799ms 561.8292 Ops/s 562.3896 Ops/s $\color{#d91a1a}-0.10\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.2025ms 25.7324ms 38.8615 Ops/s 41.2378 Ops/s $\textbf{\color{#d91a1a}-5.76\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.9043ms 0.7194ms 1.3900 KOps/s 1.3869 KOps/s $\color{#35bf28}+0.22\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7517ms 0.6770ms 1.4772 KOps/s 1.4381 KOps/s $\color{#35bf28}+2.72\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5564ms 1.4752ms 677.8586 Ops/s 678.6614 Ops/s $\color{#d91a1a}-0.12\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.9609ms 0.6958ms 1.4373 KOps/s 1.4472 KOps/s $\color{#d91a1a}-0.68\%$
test_dqn_speed 8.0850ms 1.4918ms 670.3166 Ops/s 667.2743 Ops/s $\color{#35bf28}+0.46\%$
test_ddpg_speed 4.3855ms 3.3963ms 294.4421 Ops/s 287.4405 Ops/s $\color{#35bf28}+2.44\%$
test_sac_speed 10.4132ms 9.4392ms 105.9409 Ops/s 104.9820 Ops/s $\color{#35bf28}+0.91\%$
test_redq_speed 17.2237ms 16.6052ms 60.2221 Ops/s 59.7465 Ops/s $\color{#35bf28}+0.80\%$
test_redq_deprec_speed 0.1031s 14.3621ms 69.6279 Ops/s 75.0448 Ops/s $\textbf{\color{#d91a1a}-7.22\%}$
test_td3_speed 9.7352ms 9.6019ms 104.1465 Ops/s 102.0893 Ops/s $\color{#35bf28}+2.02\%$
test_cql_speed 36.8292ms 34.6499ms 28.8601 Ops/s 28.5496 Ops/s $\color{#35bf28}+1.09\%$
test_a2c_speed 8.2994ms 7.3339ms 136.3536 Ops/s 133.3931 Ops/s $\color{#35bf28}+2.22\%$
test_ppo_speed 8.8733ms 7.6374ms 130.9346 Ops/s 128.5463 Ops/s $\color{#35bf28}+1.86\%$
test_reinforce_speed 7.6710ms 6.4507ms 155.0225 Ops/s 156.4143 Ops/s $\color{#d91a1a}-0.89\%$
test_iql_speed 29.2099ms 27.9990ms 35.7156 Ops/s 35.7360 Ops/s $\color{#d91a1a}-0.06\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.0311ms 2.4542ms 407.4643 Ops/s 401.5102 Ops/s $\color{#35bf28}+1.48\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.6860ms 2.5965ms 385.1301 Ops/s 335.1977 Ops/s $\textbf{\color{#35bf28}+14.90\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.2261ms 2.6279ms 380.5359 Ops/s 374.9521 Ops/s $\color{#35bf28}+1.49\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.0683ms 2.4592ms 406.6323 Ops/s 401.2521 Ops/s $\color{#35bf28}+1.34\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 3.4357ms 2.6194ms 381.7735 Ops/s 333.2219 Ops/s $\textbf{\color{#35bf28}+14.57\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 4.2759ms 2.6197ms 381.7287 Ops/s 375.7210 Ops/s $\color{#35bf28}+1.60\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.0917ms 2.4614ms 406.2805 Ops/s 400.9802 Ops/s $\color{#35bf28}+1.32\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.6899ms 2.6191ms 381.8133 Ops/s 332.9180 Ops/s $\textbf{\color{#35bf28}+14.69\%}$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 3.6108ms 2.6099ms 383.1513 Ops/s 373.7467 Ops/s $\color{#35bf28}+2.52\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.1096ms 2.4212ms 413.0163 Ops/s 400.4355 Ops/s $\color{#35bf28}+3.14\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.6211ms 2.6254ms 380.8923 Ops/s 331.5702 Ops/s $\textbf{\color{#35bf28}+14.88\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 3.5025ms 2.6138ms 382.5794 Ops/s 373.5629 Ops/s $\color{#35bf28}+2.41\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.1694ms 2.4316ms 411.2585 Ops/s 398.2394 Ops/s $\color{#35bf28}+3.27\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 4.1148ms 2.6374ms 379.1631 Ops/s 326.4806 Ops/s $\textbf{\color{#35bf28}+16.14\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 3.8942ms 2.6290ms 380.3690 Ops/s 372.3251 Ops/s $\color{#35bf28}+2.16\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.1916ms 2.4550ms 407.3320 Ops/s 401.1702 Ops/s $\color{#35bf28}+1.54\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.8699ms 2.6212ms 381.5066 Ops/s 374.7938 Ops/s $\color{#35bf28}+1.79\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 4.3378ms 2.6457ms 377.9657 Ops/s 374.5741 Ops/s $\color{#35bf28}+0.91\%$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1969s 18.6060ms 53.7461 Ops/s 53.2901 Ops/s $\color{#35bf28}+0.86\%$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 0.1204s 16.9817ms 58.8870 Ops/s 57.4047 Ops/s $\color{#35bf28}+2.58\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.1210s 14.8040ms 67.5494 Ops/s 57.8341 Ops/s $\textbf{\color{#35bf28}+16.80\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1195s 16.9653ms 58.9437 Ops/s 57.8664 Ops/s $\color{#35bf28}+1.86\%$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1204s 16.9524ms 58.9886 Ops/s 57.2912 Ops/s $\color{#35bf28}+2.96\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.1210s 16.8921ms 59.1992 Ops/s 65.8044 Ops/s $\textbf{\color{#d91a1a}-10.04\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1203s 16.9446ms 59.0159 Ops/s 57.6665 Ops/s $\color{#35bf28}+2.34\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.1221s 17.0507ms 58.6488 Ops/s 57.9571 Ops/s $\color{#35bf28}+1.19\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 0.1198s 16.9739ms 58.9140 Ops/s 65.7874 Ops/s $\textbf{\color{#d91a1a}-10.45\%}$

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. Tests Incomplete or broken unit tests
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants