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

[Feature] BatchSizeTransform #2030

Merged
merged 5 commits into from
Mar 26, 2024
Merged

[Feature] BatchSizeTransform #2030

merged 5 commits into from
Mar 26, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Mar 21, 2024

Copy link

pytorch-bot bot commented Mar 21, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2030

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

❌ 4 New Failures

As of commit 687647c with merge base e57d0bc (image):

NEW FAILURES - The following jobs have failed:

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 Mar 21, 2024
@vmoens vmoens added the enhancement New feature or request label Mar 21, 2024
# Conflicts:
#	torchrl/envs/transforms/transforms.py
Copy link

github-actions bot commented Mar 25, 2024

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

Total Benchmarks: 94. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 98.3868ms 97.1837ms 10.2898 Ops/s 9.7537 Ops/s $\textbf{\color{#35bf28}+5.50\%}$
test_sync 86.6367ms 86.0638ms 11.6193 Ops/s 11.6635 Ops/s $\color{#d91a1a}-0.38\%$
test_async 0.1689s 84.4599ms 11.8399 Ops/s 14.4070 Ops/s $\textbf{\color{#d91a1a}-17.82\%}$
test_single_pixels 0.1059s 0.1057s 9.4563 Ops/s 9.3191 Ops/s $\color{#35bf28}+1.47\%$
test_sync_pixels 66.6528ms 65.4421ms 15.2807 Ops/s 15.3522 Ops/s $\color{#d91a1a}-0.47\%$
test_async_pixels 0.1237s 55.7128ms 17.9492 Ops/s 18.4048 Ops/s $\color{#d91a1a}-2.48\%$
test_simple 0.7225s 0.6521s 1.5336 Ops/s 1.5454 Ops/s $\color{#d91a1a}-0.76\%$
test_transformed 0.9291s 0.8612s 1.1612 Ops/s 1.1372 Ops/s $\color{#35bf28}+2.11\%$
test_serial 2.0710s 2.0019s 0.4995 Ops/s 0.4835 Ops/s $\color{#35bf28}+3.33\%$
test_parallel 1.8202s 1.7290s 0.5784 Ops/s 0.5696 Ops/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[True-True-True-True-True] 86.3510μs 33.2796μs 30.0485 KOps/s 30.8919 KOps/s $\color{#d91a1a}-2.73\%$
test_step_mdp_speed[True-True-True-True-False] 36.6310μs 19.6238μs 50.9586 KOps/s 51.3105 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[True-True-True-False-True] 45.8210μs 18.6928μs 53.4964 KOps/s 55.3294 KOps/s $\color{#d91a1a}-3.31\%$
test_step_mdp_speed[True-True-True-False-False] 27.0300μs 11.2308μs 89.0410 KOps/s 90.9216 KOps/s $\color{#d91a1a}-2.07\%$
test_step_mdp_speed[True-True-False-True-True] 54.1210μs 34.7886μs 28.7451 KOps/s 29.1519 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[True-True-False-True-False] 47.3410μs 21.4872μs 46.5393 KOps/s 48.0765 KOps/s $\color{#d91a1a}-3.20\%$
test_step_mdp_speed[True-True-False-False-True] 36.9100μs 20.4795μs 48.8294 KOps/s 49.9345 KOps/s $\color{#d91a1a}-2.21\%$
test_step_mdp_speed[True-True-False-False-False] 0.2806ms 13.1287μs 76.1688 KOps/s 77.7136 KOps/s $\color{#d91a1a}-1.99\%$
test_step_mdp_speed[True-False-True-True-True] 60.4710μs 36.8122μs 27.1649 KOps/s 27.9652 KOps/s $\color{#d91a1a}-2.86\%$
test_step_mdp_speed[True-False-True-True-False] 47.6400μs 23.6250μs 42.3281 KOps/s 43.6983 KOps/s $\color{#d91a1a}-3.14\%$
test_step_mdp_speed[True-False-True-False-True] 36.5200μs 20.8532μs 47.9543 KOps/s 49.8558 KOps/s $\color{#d91a1a}-3.81\%$
test_step_mdp_speed[True-False-True-False-False] 34.6900μs 13.1181μs 76.2303 KOps/s 77.6609 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[True-False-False-True-True] 0.1126ms 38.5553μs 25.9368 KOps/s 26.4440 KOps/s $\color{#d91a1a}-1.92\%$
test_step_mdp_speed[True-False-False-True-False] 68.8410μs 25.4400μs 39.3081 KOps/s 40.5616 KOps/s $\color{#d91a1a}-3.09\%$
test_step_mdp_speed[True-False-False-False-True] 40.7010μs 22.3995μs 44.6438 KOps/s 45.8473 KOps/s $\color{#d91a1a}-2.63\%$
test_step_mdp_speed[True-False-False-False-False] 31.1310μs 14.9801μs 66.7550 KOps/s 68.3343 KOps/s $\color{#d91a1a}-2.31\%$
test_step_mdp_speed[False-True-True-True-True] 60.2810μs 37.0217μs 27.0112 KOps/s 27.6543 KOps/s $\color{#d91a1a}-2.33\%$
test_step_mdp_speed[False-True-True-True-False] 46.5010μs 23.6036μs 42.3665 KOps/s 43.4437 KOps/s $\color{#d91a1a}-2.48\%$
test_step_mdp_speed[False-True-True-False-True] 46.0910μs 24.8994μs 40.1616 KOps/s 41.3205 KOps/s $\color{#d91a1a}-2.80\%$
test_step_mdp_speed[False-True-True-False-False] 29.2300μs 14.8738μs 67.2323 KOps/s 69.1859 KOps/s $\color{#d91a1a}-2.82\%$
test_step_mdp_speed[False-True-False-True-True] 77.3910μs 38.7364μs 25.8155 KOps/s 25.8602 KOps/s $\color{#d91a1a}-0.17\%$
test_step_mdp_speed[False-True-False-True-False] 48.8600μs 25.4525μs 39.2889 KOps/s 39.4145 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[False-True-False-False-True] 42.0010μs 26.3261μs 37.9851 KOps/s 37.9541 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[False-True-False-False-False] 41.7410μs 16.7214μs 59.8038 KOps/s 60.3943 KOps/s $\color{#d91a1a}-0.98\%$
test_step_mdp_speed[False-False-True-True-True] 68.5410μs 40.7514μs 24.5390 KOps/s 24.6283 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[False-False-True-True-False] 45.9810μs 27.4469μs 36.4341 KOps/s 37.6204 KOps/s $\color{#d91a1a}-3.15\%$
test_step_mdp_speed[False-False-True-False-True] 48.9010μs 26.5185μs 37.7096 KOps/s 37.6924 KOps/s $\color{#35bf28}+0.05\%$
test_step_mdp_speed[False-False-True-False-False] 33.8300μs 16.7264μs 59.7857 KOps/s 60.5837 KOps/s $\color{#d91a1a}-1.32\%$
test_step_mdp_speed[False-False-False-True-True] 70.0220μs 42.3055μs 23.6376 KOps/s 24.3250 KOps/s $\color{#d91a1a}-2.83\%$
test_step_mdp_speed[False-False-False-True-False] 50.4410μs 29.2169μs 34.2268 KOps/s 35.3471 KOps/s $\color{#d91a1a}-3.17\%$
test_step_mdp_speed[False-False-False-False-True] 48.9410μs 28.0338μs 35.6712 KOps/s 36.4400 KOps/s $\color{#d91a1a}-2.11\%$
test_step_mdp_speed[False-False-False-False-False] 37.1200μs 18.5442μs 53.9253 KOps/s 54.7481 KOps/s $\color{#d91a1a}-1.50\%$
test_values[generalized_advantage_estimate-True-True] 22.0789ms 21.3261ms 46.8910 Ops/s 44.3364 Ops/s $\textbf{\color{#35bf28}+5.76\%}$
test_values[vec_generalized_advantage_estimate-True-True] 86.3456ms 3.2791ms 304.9594 Ops/s 315.0766 Ops/s $\color{#d91a1a}-3.21\%$
test_values[td0_return_estimate-False-False] 91.9110μs 60.4583μs 16.5403 KOps/s 16.1647 KOps/s $\color{#35bf28}+2.32\%$
test_values[td1_return_estimate-False-False] 49.7956ms 48.9076ms 20.4467 Ops/s 19.9414 Ops/s $\color{#35bf28}+2.53\%$
test_values[vec_td1_return_estimate-False-False] 1.9732ms 1.7329ms 577.0679 Ops/s 575.4404 Ops/s $\color{#35bf28}+0.28\%$
test_values[td_lambda_return_estimate-True-False] 78.7800ms 78.0240ms 12.8166 Ops/s 12.3786 Ops/s $\color{#35bf28}+3.54\%$
test_values[vec_td_lambda_return_estimate-True-False] 2.0388ms 1.7304ms 577.8904 Ops/s 575.5698 Ops/s $\color{#35bf28}+0.40\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 21.3783ms 21.1951ms 47.1808 Ops/s 45.1472 Ops/s $\color{#35bf28}+4.50\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.8783ms 0.6707ms 1.4910 KOps/s 1.4815 KOps/s $\color{#35bf28}+0.64\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7213ms 0.6242ms 1.6021 KOps/s 1.5973 KOps/s $\color{#35bf28}+0.30\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.4932ms 1.4284ms 700.0743 Ops/s 699.1085 Ops/s $\color{#35bf28}+0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.0103ms 0.6874ms 1.4548 KOps/s 1.5355 KOps/s $\textbf{\color{#d91a1a}-5.26\%}$
test_dqn_speed 8.0367ms 1.4641ms 683.0256 Ops/s 699.4216 Ops/s $\color{#d91a1a}-2.34\%$
test_ddpg_speed 3.0352ms 2.7426ms 364.6209 Ops/s 372.1033 Ops/s $\color{#d91a1a}-2.01\%$
test_sac_speed 8.5979ms 7.9439ms 125.8832 Ops/s 127.0896 Ops/s $\color{#d91a1a}-0.95\%$
test_redq_speed 17.7919ms 10.2686ms 97.3846 Ops/s 98.2427 Ops/s $\color{#d91a1a}-0.87\%$
test_redq_deprec_speed 11.4235ms 10.8708ms 91.9900 Ops/s 93.0421 Ops/s $\color{#d91a1a}-1.13\%$
test_td3_speed 7.9919ms 7.8527ms 127.3443 Ops/s 128.0098 Ops/s $\color{#d91a1a}-0.52\%$
test_cql_speed 26.4972ms 24.8393ms 40.2588 Ops/s 40.8086 Ops/s $\color{#d91a1a}-1.35\%$
test_a2c_speed 5.7631ms 5.5069ms 181.5905 Ops/s 187.8427 Ops/s $\color{#d91a1a}-3.33\%$
test_ppo_speed 6.2434ms 5.9261ms 168.7453 Ops/s 175.3415 Ops/s $\color{#d91a1a}-3.76\%$
test_reinforce_speed 5.1234ms 4.5226ms 221.1139 Ops/s 232.5049 Ops/s $\color{#d91a1a}-4.90\%$
test_iql_speed 20.0716ms 19.2939ms 51.8299 Ops/s 52.5077 Ops/s $\color{#d91a1a}-1.29\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.1464ms 2.9967ms 333.7003 Ops/s 349.1405 Ops/s $\color{#d91a1a}-4.42\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2014ms 0.5432ms 1.8409 KOps/s 1.8636 KOps/s $\color{#d91a1a}-1.21\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6699ms 0.5182ms 1.9297 KOps/s 1.9530 KOps/s $\color{#d91a1a}-1.19\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.3346ms 3.0136ms 331.8244 Ops/s 348.3886 Ops/s $\color{#d91a1a}-4.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.1223s 0.6502ms 1.5380 KOps/s 1.8900 KOps/s $\textbf{\color{#d91a1a}-18.63\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6838ms 0.5121ms 1.9526 KOps/s 1.9752 KOps/s $\color{#d91a1a}-1.14\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 4.2089ms 1.3933ms 717.7252 Ops/s 722.7399 Ops/s $\color{#d91a1a}-0.69\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5186ms 1.3195ms 757.8635 Ops/s 757.5056 Ops/s $\color{#35bf28}+0.05\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.5591ms 3.1192ms 320.5939 Ops/s 334.9536 Ops/s $\color{#d91a1a}-4.29\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8198ms 0.6650ms 1.5038 KOps/s 1.5136 KOps/s $\color{#d91a1a}-0.65\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 5.8811ms 0.6501ms 1.5383 KOps/s 1.5555 KOps/s $\color{#d91a1a}-1.11\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.1784ms 3.0070ms 332.5623 Ops/s 350.4793 Ops/s $\textbf{\color{#d91a1a}-5.11\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.5875ms 0.5468ms 1.8287 KOps/s 1.8577 KOps/s $\color{#d91a1a}-1.56\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6865ms 0.5189ms 1.9272 KOps/s 1.9433 KOps/s $\color{#d91a1a}-0.83\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.2409ms 3.0527ms 327.5789 Ops/s 349.4489 Ops/s $\textbf{\color{#d91a1a}-6.26\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.2283ms 0.5407ms 1.8495 KOps/s 1.8819 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7233ms 0.5151ms 1.9414 KOps/s 1.9755 KOps/s $\color{#d91a1a}-1.73\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.2718ms 3.1414ms 318.3320 Ops/s 334.1695 Ops/s $\color{#d91a1a}-4.74\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8173ms 0.6718ms 1.4885 KOps/s 1.5096 KOps/s $\color{#d91a1a}-1.40\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7615ms 0.6484ms 1.5423 KOps/s 1.5593 KOps/s $\color{#d91a1a}-1.09\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1372s 7.6478ms 130.7574 Ops/s 138.7468 Ops/s $\textbf{\color{#d91a1a}-5.76\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 17.1664ms 14.7230ms 67.9209 Ops/s 67.4706 Ops/s $\color{#35bf28}+0.67\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.1441ms 1.0313ms 969.6228 Ops/s 934.2267 Ops/s $\color{#35bf28}+3.79\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1145s 7.0019ms 142.8187 Ops/s 106.4061 Ops/s $\textbf{\color{#35bf28}+34.22\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1273s 16.7221ms 59.8011 Ops/s 67.9551 Ops/s $\textbf{\color{#d91a1a}-12.00\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.0801ms 0.9952ms 1.0048 KOps/s 937.7059 Ops/s $\textbf{\color{#35bf28}+7.16\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1168s 7.3440ms 136.1649 Ops/s 134.5591 Ops/s $\color{#35bf28}+1.19\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 17.2623ms 14.8083ms 67.5297 Ops/s 66.4834 Ops/s $\color{#35bf28}+1.57\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.6411ms 1.4237ms 702.4132 Ops/s 719.5333 Ops/s $\color{#d91a1a}-2.38\%$

Copy link

github-actions bot commented Mar 25, 2024

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

Total Benchmarks: 91. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 54.3489ms 53.2906ms 18.7651 Ops/s 18.0633 Ops/s $\color{#35bf28}+3.88\%$
test_sync 40.9002ms 30.1839ms 33.1303 Ops/s 35.5799 Ops/s $\textbf{\color{#d91a1a}-6.89\%}$
test_async 43.7192ms 28.7981ms 34.7245 Ops/s 37.2473 Ops/s $\textbf{\color{#d91a1a}-6.77\%}$
test_simple 0.4130s 0.3487s 2.8681 Ops/s 2.8978 Ops/s $\color{#d91a1a}-1.02\%$
test_transformed 0.5597s 0.4952s 2.0195 Ops/s 2.0684 Ops/s $\color{#d91a1a}-2.36\%$
test_serial 1.2905s 1.2274s 0.8147 Ops/s 0.8202 Ops/s $\color{#d91a1a}-0.66\%$
test_parallel 1.0616s 0.9987s 1.0013 Ops/s 1.0095 Ops/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[True-True-True-True-True] 0.1615ms 22.0870μs 45.2756 KOps/s 45.6377 KOps/s $\color{#d91a1a}-0.79\%$
test_step_mdp_speed[True-True-True-True-False] 42.3590μs 13.1931μs 75.7971 KOps/s 75.8922 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-True-True-False-True] 32.3800μs 12.6357μs 79.1408 KOps/s 78.1261 KOps/s $\color{#35bf28}+1.30\%$
test_step_mdp_speed[True-True-True-False-False] 35.0950μs 7.6467μs 130.7755 KOps/s 129.7103 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-True-False-True-True] 60.2020μs 23.4060μs 42.7240 KOps/s 42.8477 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[True-True-False-True-False] 43.7120μs 14.5251μs 68.8461 KOps/s 68.7048 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[True-True-False-False-True] 62.0670μs 14.0207μs 71.3233 KOps/s 70.8044 KOps/s $\color{#35bf28}+0.73\%$
test_step_mdp_speed[True-True-False-False-False] 41.5680μs 8.8605μs 112.8605 KOps/s 111.3717 KOps/s $\color{#35bf28}+1.34\%$
test_step_mdp_speed[True-False-True-True-True] 84.5880μs 24.5412μs 40.7478 KOps/s 40.3262 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-False-True-True-False] 58.3690μs 15.9409μs 62.7317 KOps/s 63.1120 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[True-False-True-False-True] 67.9470μs 13.9562μs 71.6529 KOps/s 70.1240 KOps/s $\color{#35bf28}+2.18\%$
test_step_mdp_speed[True-False-True-False-False] 54.8630μs 8.8986μs 112.3766 KOps/s 111.9519 KOps/s $\color{#35bf28}+0.38\%$
test_step_mdp_speed[True-False-False-True-True] 69.6200μs 25.8802μs 38.6396 KOps/s 38.7600 KOps/s $\color{#d91a1a}-0.31\%$
test_step_mdp_speed[True-False-False-True-False] 60.5830μs 17.3372μs 57.6793 KOps/s 58.7172 KOps/s $\color{#d91a1a}-1.77\%$
test_step_mdp_speed[True-False-False-False-True] 77.9340μs 15.2111μs 65.7414 KOps/s 65.8228 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-False-False-False-False] 37.8610μs 10.1978μs 98.0607 KOps/s 98.4287 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[False-True-True-True-True] 62.6370μs 24.7197μs 40.4535 KOps/s 40.6872 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[False-True-True-True-False] 61.1140μs 15.9759μs 62.5944 KOps/s 63.3915 KOps/s $\color{#d91a1a}-1.26\%$
test_step_mdp_speed[False-True-True-False-True] 81.5600μs 16.2286μs 61.6195 KOps/s 61.4861 KOps/s $\color{#35bf28}+0.22\%$
test_step_mdp_speed[False-True-True-False-False] 38.2610μs 10.2494μs 97.5666 KOps/s 98.8304 KOps/s $\color{#d91a1a}-1.28\%$
test_step_mdp_speed[False-True-False-True-True] 50.4950μs 26.5099μs 37.7218 KOps/s 38.1974 KOps/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[False-True-False-True-False] 85.3390μs 17.2853μs 57.8525 KOps/s 58.4288 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[False-True-False-False-True] 44.6430μs 17.6726μs 56.5846 KOps/s 56.6607 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[False-True-False-False-False] 41.1270μs 11.5635μs 86.4793 KOps/s 87.6510 KOps/s $\color{#d91a1a}-1.34\%$
test_step_mdp_speed[False-False-True-True-True] 69.9410μs 27.3627μs 36.5461 KOps/s 36.7102 KOps/s $\color{#d91a1a}-0.45\%$
test_step_mdp_speed[False-False-True-True-False] 44.7840μs 18.7020μs 53.4702 KOps/s 54.1544 KOps/s $\color{#d91a1a}-1.26\%$
test_step_mdp_speed[False-False-True-False-True] 43.6220μs 17.6940μs 56.5162 KOps/s 57.2646 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[False-False-True-False-False] 35.6060μs 11.6593μs 85.7684 KOps/s 87.3743 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[False-False-False-True-True] 60.1620μs 28.6839μs 34.8627 KOps/s 35.3839 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-False-False-True-False] 64.3220μs 19.8187μs 50.4573 KOps/s 51.4295 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[False-False-False-False-True] 76.8730μs 18.7257μs 53.4026 KOps/s 53.7347 KOps/s $\color{#d91a1a}-0.62\%$
test_step_mdp_speed[False-False-False-False-False] 42.5200μs 12.6593μs 78.9930 KOps/s 79.6508 KOps/s $\color{#d91a1a}-0.83\%$
test_values[generalized_advantage_estimate-True-True] 10.4256ms 9.4290ms 106.0560 Ops/s 107.8991 Ops/s $\color{#d91a1a}-1.71\%$
test_values[vec_generalized_advantage_estimate-True-True] 36.9142ms 33.8620ms 29.5317 Ops/s 28.1802 Ops/s $\color{#35bf28}+4.80\%$
test_values[td0_return_estimate-False-False] 0.2307ms 0.1741ms 5.7455 KOps/s 5.5499 KOps/s $\color{#35bf28}+3.52\%$
test_values[td1_return_estimate-False-False] 30.2335ms 23.1795ms 43.1416 Ops/s 43.0435 Ops/s $\color{#35bf28}+0.23\%$
test_values[vec_td1_return_estimate-False-False] 35.4062ms 33.8798ms 29.5162 Ops/s 28.0275 Ops/s $\textbf{\color{#35bf28}+5.31\%}$
test_values[td_lambda_return_estimate-True-False] 35.9587ms 33.4995ms 29.8512 Ops/s 29.6812 Ops/s $\color{#35bf28}+0.57\%$
test_values[vec_td_lambda_return_estimate-True-False] 39.0295ms 34.2691ms 29.1808 Ops/s 28.1194 Ops/s $\color{#35bf28}+3.77\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.3094ms 8.0445ms 124.3086 Ops/s 122.0121 Ops/s $\color{#35bf28}+1.88\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3787ms 1.9596ms 510.3113 Ops/s 511.4659 Ops/s $\color{#d91a1a}-0.23\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5219ms 0.3537ms 2.8276 KOps/s 2.8214 KOps/s $\color{#35bf28}+0.22\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 40.5009ms 39.5251ms 25.3004 Ops/s 21.7989 Ops/s $\textbf{\color{#35bf28}+16.06\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.6823ms 3.1575ms 316.7056 Ops/s 330.9819 Ops/s $\color{#d91a1a}-4.31\%$
test_dqn_speed 7.4214ms 1.3878ms 720.5771 Ops/s 736.3747 Ops/s $\color{#d91a1a}-2.15\%$
test_ddpg_speed 3.2232ms 2.7603ms 362.2771 Ops/s 370.1300 Ops/s $\color{#d91a1a}-2.12\%$
test_sac_speed 9.9532ms 8.3794ms 119.3405 Ops/s 112.8927 Ops/s $\textbf{\color{#35bf28}+5.71\%}$
test_redq_speed 14.4512ms 13.4713ms 74.2321 Ops/s 75.8377 Ops/s $\color{#d91a1a}-2.12\%$
test_redq_deprec_speed 97.8147ms 14.8107ms 67.5187 Ops/s 76.7659 Ops/s $\textbf{\color{#d91a1a}-12.05\%}$
test_td3_speed 8.7824ms 8.2575ms 121.1016 Ops/s 121.4695 Ops/s $\color{#d91a1a}-0.30\%$
test_cql_speed 38.0220ms 36.4904ms 27.4045 Ops/s 27.2986 Ops/s $\color{#35bf28}+0.39\%$
test_a2c_speed 8.1000ms 7.4198ms 134.7737 Ops/s 134.8758 Ops/s $\color{#d91a1a}-0.08\%$
test_ppo_speed 9.1600ms 7.7091ms 129.7170 Ops/s 129.4913 Ops/s $\color{#35bf28}+0.17\%$
test_reinforce_speed 7.1727ms 6.5920ms 151.6979 Ops/s 151.4866 Ops/s $\color{#35bf28}+0.14\%$
test_iql_speed 33.8263ms 32.6830ms 30.5969 Ops/s 30.5811 Ops/s $\color{#35bf28}+0.05\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.5546ms 2.2996ms 434.8654 Ops/s 422.2930 Ops/s $\color{#35bf28}+2.98\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2265ms 0.5063ms 1.9752 KOps/s 1.9282 KOps/s $\color{#35bf28}+2.44\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7496ms 0.4787ms 2.0890 KOps/s 2.0636 KOps/s $\color{#35bf28}+1.23\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.4061ms 2.3181ms 431.3952 Ops/s 407.9908 Ops/s $\textbf{\color{#35bf28}+5.74\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1101ms 0.4992ms 2.0032 KOps/s 1.9550 KOps/s $\color{#35bf28}+2.46\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6756ms 0.4753ms 2.1041 KOps/s 2.0755 KOps/s $\color{#35bf28}+1.38\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.3180ms 1.2145ms 823.4120 Ops/s 799.6965 Ops/s $\color{#35bf28}+2.97\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.3443ms 1.1444ms 873.7840 Ops/s 855.8672 Ops/s $\color{#35bf28}+2.09\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.5539ms 2.3825ms 419.7297 Ops/s 396.7025 Ops/s $\textbf{\color{#35bf28}+5.80\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1128ms 0.6199ms 1.6132 KOps/s 1.5746 KOps/s $\color{#35bf28}+2.45\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7776ms 0.5934ms 1.6853 KOps/s 1.6625 KOps/s $\color{#35bf28}+1.37\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.6219ms 2.2931ms 436.0963 Ops/s 424.6391 Ops/s $\color{#35bf28}+2.70\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0391ms 0.5057ms 1.9776 KOps/s 1.9206 KOps/s $\color{#35bf28}+2.96\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6300ms 0.4795ms 2.0854 KOps/s 2.0055 KOps/s $\color{#35bf28}+3.98\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.6041ms 2.2925ms 436.1981 Ops/s 411.8752 Ops/s $\textbf{\color{#35bf28}+5.91\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6595ms 0.5019ms 1.9924 KOps/s 1.9598 KOps/s $\color{#35bf28}+1.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 3.6623ms 0.4802ms 2.0823 KOps/s 2.0664 KOps/s $\color{#35bf28}+0.77\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.7939ms 2.3998ms 416.6977 Ops/s 416.6106 Ops/s $\color{#35bf28}+0.02\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1637ms 0.6239ms 1.6028 KOps/s 1.5660 KOps/s $\color{#35bf28}+2.35\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7428ms 0.5918ms 1.6896 KOps/s 1.6398 KOps/s $\color{#35bf28}+3.04\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1148s 7.9929ms 125.1110 Ops/s 117.9332 Ops/s $\textbf{\color{#35bf28}+6.09\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 14.4695ms 12.2056ms 81.9296 Ops/s 82.0594 Ops/s $\color{#d91a1a}-0.16\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.6069ms 1.0733ms 931.7247 Ops/s 918.2486 Ops/s $\color{#35bf28}+1.47\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1050s 5.7339ms 174.4002 Ops/s 168.7127 Ops/s $\color{#35bf28}+3.37\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 14.4824ms 12.2556ms 81.5954 Ops/s 81.8487 Ops/s $\color{#d91a1a}-0.31\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 4.0071ms 1.1308ms 884.2996 Ops/s 912.2367 Ops/s $\color{#d91a1a}-3.06\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1082s 8.3316ms 120.0249 Ops/s 114.5555 Ops/s $\color{#35bf28}+4.77\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 14.9929ms 12.5586ms 79.6266 Ops/s 72.0937 Ops/s $\textbf{\color{#35bf28}+10.45\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 4.4112ms 1.4558ms 686.9084 Ops/s 663.3291 Ops/s $\color{#35bf28}+3.55\%$

@vmoens vmoens merged commit a7bf5a4 into main Mar 26, 2024
63 of 67 checks passed
@vmoens vmoens deleted the batch-size-transform branch March 26, 2024 19:28
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. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants