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] Fix replay buffer extension with lists #1937

Merged
merged 2 commits into from
Feb 20, 2024
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 20, 2024

Fixes #1934

Copy link

pytorch-bot bot commented Feb 20, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1937

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

✅ You can merge normally! (3 Unrelated Failures)

As of commit 8961a04 with merge base 78b31a9 (image):

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

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

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

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 20, 2024
@vmoens vmoens added bug Something isn't working Suitable for minor Suitable to be integrated in minor release (no new feature) labels Feb 20, 2024
Copy link

github-actions bot commented Feb 20, 2024

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}3$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 63.5724ms 62.8974ms 15.8989 Ops/s 15.0370 Ops/s $\textbf{\color{#35bf28}+5.73\%}$
test_sync 55.8771ms 40.0102ms 24.9936 Ops/s 29.1461 Ops/s $\textbf{\color{#d91a1a}-14.25\%}$
test_async 58.1784ms 31.9971ms 31.2528 Ops/s 31.6580 Ops/s $\color{#d91a1a}-1.28\%$
test_simple 0.4997s 0.4396s 2.2749 Ops/s 2.2472 Ops/s $\color{#35bf28}+1.23\%$
test_transformed 0.6511s 0.6024s 1.6600 Ops/s 1.6538 Ops/s $\color{#35bf28}+0.38\%$
test_serial 1.4971s 1.4454s 0.6918 Ops/s 0.6840 Ops/s $\color{#35bf28}+1.16\%$
test_parallel 1.4485s 1.3856s 0.7217 Ops/s 0.7202 Ops/s $\color{#35bf28}+0.20\%$
test_step_mdp_speed[True-True-True-True-True] 0.2565ms 21.2249μs 47.1145 KOps/s 46.4005 KOps/s $\color{#35bf28}+1.54\%$
test_step_mdp_speed[True-True-True-True-False] 66.0130μs 12.6018μs 79.3538 KOps/s 77.8683 KOps/s $\color{#35bf28}+1.91\%$
test_step_mdp_speed[True-True-True-False-True] 38.0710μs 12.2847μs 81.4017 KOps/s 80.2635 KOps/s $\color{#35bf28}+1.42\%$
test_step_mdp_speed[True-True-True-False-False] 48.2900μs 7.3235μs 136.5466 KOps/s 133.5377 KOps/s $\color{#35bf28}+2.25\%$
test_step_mdp_speed[True-True-False-True-True] 64.5600μs 22.2152μs 45.0142 KOps/s 44.0351 KOps/s $\color{#35bf28}+2.22\%$
test_step_mdp_speed[True-True-False-True-False] 58.1580μs 13.8647μs 72.1255 KOps/s 70.4808 KOps/s $\color{#35bf28}+2.33\%$
test_step_mdp_speed[True-True-False-False-True] 48.3000μs 13.4125μs 74.5571 KOps/s 73.3879 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-True-False-False-False] 48.3800μs 8.6526μs 115.5717 KOps/s 115.3159 KOps/s $\color{#35bf28}+0.22\%$
test_step_mdp_speed[True-False-True-True-True] 58.2690μs 23.6294μs 42.3201 KOps/s 41.3742 KOps/s $\color{#35bf28}+2.29\%$
test_step_mdp_speed[True-False-True-True-False] 50.2040μs 15.3809μs 65.0159 KOps/s 64.7629 KOps/s $\color{#35bf28}+0.39\%$
test_step_mdp_speed[True-False-True-False-True] 39.0130μs 13.3456μs 74.9308 KOps/s 72.5382 KOps/s $\color{#35bf28}+3.30\%$
test_step_mdp_speed[True-False-True-False-False] 41.0570μs 8.6566μs 115.5187 KOps/s 113.9895 KOps/s $\color{#35bf28}+1.34\%$
test_step_mdp_speed[True-False-False-True-True] 98.9050μs 24.7450μs 40.4123 KOps/s 39.5099 KOps/s $\color{#35bf28}+2.28\%$
test_step_mdp_speed[True-False-False-True-False] 47.4680μs 16.3840μs 61.0353 KOps/s 59.9181 KOps/s $\color{#35bf28}+1.86\%$
test_step_mdp_speed[True-False-False-False-True] 58.7890μs 14.5447μs 68.7537 KOps/s 67.3023 KOps/s $\color{#35bf28}+2.16\%$
test_step_mdp_speed[True-False-False-False-False] 41.8580μs 9.7978μs 102.0636 KOps/s 101.2292 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-True-True-True-True] 56.8970μs 23.6479μs 42.2870 KOps/s 41.2617 KOps/s $\color{#35bf28}+2.48\%$
test_step_mdp_speed[False-True-True-True-False] 44.7140μs 15.2317μs 65.6526 KOps/s 64.7299 KOps/s $\color{#35bf28}+1.43\%$
test_step_mdp_speed[False-True-True-False-True] 54.6920μs 15.6456μs 63.9156 KOps/s 63.4810 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[False-True-True-False-False] 41.4870μs 9.8544μs 101.4771 KOps/s 101.1429 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[False-True-False-True-True] 43.3410μs 25.1420μs 39.7741 KOps/s 38.9297 KOps/s $\color{#35bf28}+2.17\%$
test_step_mdp_speed[False-True-False-True-False] 62.4160μs 16.4090μs 60.9420 KOps/s 60.0085 KOps/s $\color{#35bf28}+1.56\%$
test_step_mdp_speed[False-True-False-False-True] 51.6060μs 16.7828μs 59.5847 KOps/s 58.6659 KOps/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[False-True-False-False-False] 36.1270μs 11.1030μs 90.0659 KOps/s 90.1709 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-False-True-True-True] 86.9020μs 26.2094μs 38.1543 KOps/s 37.6626 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[False-False-True-True-False] 50.6240μs 17.6688μs 56.5970 KOps/s 55.6566 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[False-False-True-False-True] 56.3450μs 16.7848μs 59.5778 KOps/s 58.2694 KOps/s $\color{#35bf28}+2.25\%$
test_step_mdp_speed[False-False-True-False-False] 33.7330μs 11.0030μs 90.8842 KOps/s 89.3991 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[False-False-False-True-True] 73.1660μs 26.7213μs 37.4233 KOps/s 36.0893 KOps/s $\color{#35bf28}+3.70\%$
test_step_mdp_speed[False-False-False-True-False] 47.8090μs 18.7756μs 53.2606 KOps/s 52.5067 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[False-False-False-False-True] 58.1490μs 17.7473μs 56.3466 KOps/s 55.7267 KOps/s $\color{#35bf28}+1.11\%$
test_step_mdp_speed[False-False-False-False-False] 54.3710μs 12.1127μs 82.5578 KOps/s 82.2745 KOps/s $\color{#35bf28}+0.34\%$
test_values[generalized_advantage_estimate-True-True] 12.1466ms 9.3821ms 106.5854 Ops/s 104.8315 Ops/s $\color{#35bf28}+1.67\%$
test_values[vec_generalized_advantage_estimate-True-True] 37.9750ms 33.5766ms 29.7827 Ops/s 29.7690 Ops/s $\color{#35bf28}+0.05\%$
test_values[td0_return_estimate-False-False] 0.2411ms 0.1772ms 5.6448 KOps/s 5.5279 KOps/s $\color{#35bf28}+2.11\%$
test_values[td1_return_estimate-False-False] 26.9922ms 23.8670ms 41.8988 Ops/s 42.1521 Ops/s $\color{#d91a1a}-0.60\%$
test_values[vec_td1_return_estimate-False-False] 34.8336ms 33.5792ms 29.7803 Ops/s 29.7240 Ops/s $\color{#35bf28}+0.19\%$
test_values[td_lambda_return_estimate-True-False] 37.1192ms 34.0780ms 29.3445 Ops/s 29.2904 Ops/s $\color{#35bf28}+0.18\%$
test_values[vec_td_lambda_return_estimate-True-False] 34.6703ms 33.4247ms 29.9180 Ops/s 29.6721 Ops/s $\color{#35bf28}+0.83\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 16.9476ms 8.5416ms 117.0739 Ops/s 121.9214 Ops/s $\color{#d91a1a}-3.98\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.2851ms 1.9948ms 501.3001 Ops/s 494.8969 Ops/s $\color{#35bf28}+1.29\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.9736ms 0.3588ms 2.7871 KOps/s 2.8475 KOps/s $\color{#d91a1a}-2.12\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 40.1384ms 39.5476ms 25.2860 Ops/s 24.5580 Ops/s $\color{#35bf28}+2.96\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.8769ms 3.0734ms 325.3686 Ops/s 326.2771 Ops/s $\color{#d91a1a}-0.28\%$
test_dqn_speed 71.9037ms 1.4678ms 681.2733 Ops/s 719.1079 Ops/s $\textbf{\color{#d91a1a}-5.26\%}$
test_ddpg_speed 3.4593ms 2.7267ms 366.7413 Ops/s 362.2411 Ops/s $\color{#35bf28}+1.24\%$
test_sac_speed 9.3868ms 8.7854ms 113.8249 Ops/s 111.4544 Ops/s $\color{#35bf28}+2.13\%$
test_redq_speed 15.8186ms 13.9113ms 71.8841 Ops/s 71.5724 Ops/s $\color{#35bf28}+0.44\%$
test_redq_deprec_speed 15.4293ms 14.5730ms 68.6202 Ops/s 68.7465 Ops/s $\color{#d91a1a}-0.18\%$
test_td3_speed 9.4708ms 8.9842ms 111.3063 Ops/s 110.0083 Ops/s $\color{#35bf28}+1.18\%$
test_cql_speed 38.4790ms 37.6221ms 26.5802 Ops/s 25.9504 Ops/s $\color{#35bf28}+2.43\%$
test_a2c_speed 9.0390ms 7.7666ms 128.7570 Ops/s 127.4734 Ops/s $\color{#35bf28}+1.01\%$
test_ppo_speed 8.7937ms 8.1130ms 123.2592 Ops/s 121.8875 Ops/s $\color{#35bf28}+1.13\%$
test_reinforce_speed 8.1912ms 6.8883ms 145.1739 Ops/s 143.6455 Ops/s $\color{#35bf28}+1.06\%$
test_iql_speed 35.5342ms 33.8794ms 29.5165 Ops/s 29.2274 Ops/s $\color{#35bf28}+0.99\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.7312ms 3.0786ms 324.8187 Ops/s 320.9715 Ops/s $\color{#35bf28}+1.20\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6506ms 0.5279ms 1.8944 KOps/s 1.8040 KOps/s $\textbf{\color{#35bf28}+5.01\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7387ms 0.4964ms 2.0144 KOps/s 2.0044 KOps/s $\color{#35bf28}+0.50\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 4.3520ms 2.9913ms 334.3005 Ops/s 327.6026 Ops/s $\color{#35bf28}+2.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8185ms 0.5273ms 1.8964 KOps/s 1.9071 KOps/s $\color{#d91a1a}-0.56\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8594ms 0.5067ms 1.9735 KOps/s 2.0044 KOps/s $\color{#d91a1a}-1.54\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 4.0723ms 3.0889ms 323.7420 Ops/s 313.0974 Ops/s $\color{#35bf28}+3.40\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7892ms 0.6457ms 1.5488 KOps/s 1.4955 KOps/s $\color{#35bf28}+3.56\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9457ms 0.6166ms 1.6218 KOps/s 1.6121 KOps/s $\color{#35bf28}+0.60\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.2407ms 2.9614ms 337.6834 Ops/s 319.6455 Ops/s $\textbf{\color{#35bf28}+5.64\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9098ms 0.5751ms 1.7390 KOps/s 1.8935 KOps/s $\textbf{\color{#d91a1a}-8.16\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6048ms 0.5046ms 1.9819 KOps/s 1.9804 KOps/s $\color{#35bf28}+0.08\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.2770ms 2.9854ms 334.9690 Ops/s 330.6664 Ops/s $\color{#35bf28}+1.30\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9159ms 0.5341ms 1.8722 KOps/s 1.9016 KOps/s $\color{#d91a1a}-1.54\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6293ms 0.4974ms 2.0104 KOps/s 1.9785 KOps/s $\color{#35bf28}+1.61\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 4.5126ms 3.1053ms 322.0347 Ops/s 312.5755 Ops/s $\color{#35bf28}+3.03\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9122ms 0.6474ms 1.5447 KOps/s 1.5410 KOps/s $\color{#35bf28}+0.24\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0026ms 0.6181ms 1.6177 KOps/s 1.6139 KOps/s $\color{#35bf28}+0.24\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1169s 8.3214ms 120.1723 Ops/s 119.4851 Ops/s $\color{#35bf28}+0.58\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 16.0769ms 13.4690ms 74.2444 Ops/s 75.1520 Ops/s $\color{#d91a1a}-1.21\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.3255ms 2.6033ms 384.1211 Ops/s 389.9008 Ops/s $\color{#d91a1a}-1.48\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1031s 9.9250ms 100.7552 Ops/s 99.3968 Ops/s $\color{#35bf28}+1.37\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 17.5899ms 13.5082ms 74.0291 Ops/s 74.2237 Ops/s $\color{#d91a1a}-0.26\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 4.9626ms 2.5601ms 390.6094 Ops/s 385.6214 Ops/s $\color{#35bf28}+1.29\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1025s 10.2461ms 97.5983 Ops/s 95.2585 Ops/s $\color{#35bf28}+2.46\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 16.6291ms 13.7402ms 72.7794 Ops/s 72.5331 Ops/s $\color{#35bf28}+0.34\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.2610ms 2.9368ms 340.5107 Ops/s 347.6636 Ops/s $\color{#d91a1a}-2.06\%$

Copy link

github-actions bot commented Feb 20, 2024

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

Total Benchmarks: 92. 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 0.1131s 0.1128s 8.8684 Ops/s 8.4686 Ops/s $\color{#35bf28}+4.72\%$
test_sync 95.3914ms 95.0650ms 10.5191 Ops/s 10.4611 Ops/s $\color{#35bf28}+0.55\%$
test_async 0.1814s 91.8895ms 10.8826 Ops/s 10.9695 Ops/s $\color{#d91a1a}-0.79\%$
test_single_pixels 0.2030s 0.1418s 7.0509 Ops/s 8.0400 Ops/s $\textbf{\color{#d91a1a}-12.30\%}$
test_sync_pixels 83.8452ms 82.5985ms 12.1068 Ops/s 12.5569 Ops/s $\color{#d91a1a}-3.59\%$
test_async_pixels 0.1441s 73.2673ms 13.6487 Ops/s 13.2039 Ops/s $\color{#35bf28}+3.37\%$
test_simple 0.8089s 0.8080s 1.2377 Ops/s 1.2243 Ops/s $\color{#35bf28}+1.09\%$
test_transformed 1.0382s 1.0377s 0.9637 Ops/s 0.9579 Ops/s $\color{#35bf28}+0.60\%$
test_serial 2.5059s 2.4482s 0.4085 Ops/s 0.4111 Ops/s $\color{#d91a1a}-0.64\%$
test_parallel 2.1396s 2.1030s 0.4755 Ops/s 0.4868 Ops/s $\color{#d91a1a}-2.32\%$
test_step_mdp_speed[True-True-True-True-True] 0.1120ms 32.7749μs 30.5112 KOps/s 30.1299 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[True-True-True-True-False] 0.1410ms 19.7460μs 50.6432 KOps/s 51.0431 KOps/s $\color{#d91a1a}-0.78\%$
test_step_mdp_speed[True-True-True-False-True] 0.1657ms 18.7030μs 53.4672 KOps/s 53.6744 KOps/s $\color{#d91a1a}-0.39\%$
test_step_mdp_speed[True-True-True-False-False] 26.8010μs 11.1271μs 89.8704 KOps/s 89.4999 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-True-False-True-True] 81.5240μs 34.9199μs 28.6369 KOps/s 28.9378 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[True-True-False-True-False] 39.9420μs 21.5971μs 46.3024 KOps/s 46.3031 KOps/s $-0.00\%$
test_step_mdp_speed[True-True-False-False-True] 61.7430μs 20.5487μs 48.6649 KOps/s 48.7983 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[True-True-False-False-False] 38.3520μs 13.0066μs 76.8839 KOps/s 75.6865 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[True-False-True-True-True] 54.8520μs 36.9516μs 27.0624 KOps/s 26.9287 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[True-False-True-True-False] 43.2020μs 23.3387μs 42.8474 KOps/s 42.5833 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[True-False-True-False-True] 44.5630μs 20.3813μs 49.0645 KOps/s 49.0188 KOps/s $\color{#35bf28}+0.09\%$
test_step_mdp_speed[True-False-True-False-False] 33.9210μs 13.0922μs 76.3816 KOps/s 75.2847 KOps/s $\color{#35bf28}+1.46\%$
test_step_mdp_speed[True-False-False-True-True] 59.8230μs 38.4440μs 26.0119 KOps/s 25.5896 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-False-False-True-False] 0.2170ms 25.3305μs 39.4781 KOps/s 39.3653 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[True-False-False-False-True] 39.7110μs 22.4071μs 44.6288 KOps/s 45.4961 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[True-False-False-False-False] 29.9620μs 14.7846μs 67.6381 KOps/s 66.6219 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-True-True-True-True] 62.2130μs 36.4335μs 27.4473 KOps/s 27.1824 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[False-True-True-True-False] 41.6020μs 23.6588μs 42.2675 KOps/s 42.6956 KOps/s $\color{#d91a1a}-1.00\%$
test_step_mdp_speed[False-True-True-False-True] 50.4230μs 24.2357μs 41.2614 KOps/s 40.6482 KOps/s $\color{#35bf28}+1.51\%$
test_step_mdp_speed[False-True-True-False-False] 30.7710μs 14.8283μs 67.4387 KOps/s 66.7328 KOps/s $\color{#35bf28}+1.06\%$
test_step_mdp_speed[False-True-False-True-True] 81.6840μs 38.4560μs 26.0038 KOps/s 25.6354 KOps/s $\color{#35bf28}+1.44\%$
test_step_mdp_speed[False-True-False-True-False] 44.9930μs 25.4686μs 39.2640 KOps/s 39.0527 KOps/s $\color{#35bf28}+0.54\%$
test_step_mdp_speed[False-True-False-False-True] 41.9420μs 26.3383μs 37.9675 KOps/s 37.9841 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[False-True-False-False-False] 31.0810μs 16.7438μs 59.7235 KOps/s 59.8567 KOps/s $\color{#d91a1a}-0.22\%$
test_step_mdp_speed[False-False-True-True-True] 72.1540μs 40.2200μs 24.8632 KOps/s 24.7980 KOps/s $\color{#35bf28}+0.26\%$
test_step_mdp_speed[False-False-True-True-False] 45.5620μs 27.4922μs 36.3740 KOps/s 36.5150 KOps/s $\color{#d91a1a}-0.39\%$
test_step_mdp_speed[False-False-True-False-True] 42.4420μs 26.2391μs 38.1110 KOps/s 38.1874 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[False-False-True-False-False] 37.6310μs 16.7458μs 59.7164 KOps/s 59.2016 KOps/s $\color{#35bf28}+0.87\%$
test_step_mdp_speed[False-False-False-True-True] 68.6940μs 41.8831μs 23.8760 KOps/s 23.7422 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-False-False-True-False] 52.7930μs 28.8173μs 34.7014 KOps/s 34.4525 KOps/s $\color{#35bf28}+0.72\%$
test_step_mdp_speed[False-False-False-False-True] 59.9630μs 27.6396μs 36.1800 KOps/s 35.7711 KOps/s $\color{#35bf28}+1.14\%$
test_step_mdp_speed[False-False-False-False-False] 32.4320μs 18.4231μs 54.2798 KOps/s 54.1315 KOps/s $\color{#35bf28}+0.27\%$
test_values[generalized_advantage_estimate-True-True] 24.6753ms 24.2394ms 41.2552 Ops/s 39.9394 Ops/s $\color{#35bf28}+3.29\%$
test_values[vec_generalized_advantage_estimate-True-True] 83.2971ms 3.2232ms 310.2483 Ops/s 310.7337 Ops/s $\color{#d91a1a}-0.16\%$
test_values[td0_return_estimate-False-False] 99.0250μs 59.4644μs 16.8168 KOps/s 17.2989 KOps/s $\color{#d91a1a}-2.79\%$
test_values[td1_return_estimate-False-False] 51.9144ms 51.5386ms 19.4029 Ops/s 19.6993 Ops/s $\color{#d91a1a}-1.50\%$
test_values[vec_td1_return_estimate-False-False] 2.0834ms 1.7515ms 570.9372 Ops/s 573.0319 Ops/s $\color{#d91a1a}-0.37\%$
test_values[td_lambda_return_estimate-True-False] 83.1412ms 82.7015ms 12.0917 Ops/s 12.2859 Ops/s $\color{#d91a1a}-1.58\%$
test_values[vec_td_lambda_return_estimate-True-False] 3.9259ms 1.7846ms 560.3606 Ops/s 561.3487 Ops/s $\color{#d91a1a}-0.18\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 23.1110ms 22.8338ms 43.7947 Ops/s 44.6173 Ops/s $\color{#d91a1a}-1.84\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.8693ms 0.6903ms 1.4487 KOps/s 1.4432 KOps/s $\color{#35bf28}+0.38\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6852ms 0.6399ms 1.5628 KOps/s 1.5369 KOps/s $\color{#35bf28}+1.69\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6018ms 1.4446ms 692.2235 Ops/s 696.4400 Ops/s $\color{#d91a1a}-0.61\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.9424ms 0.6615ms 1.5117 KOps/s 1.5343 KOps/s $\color{#d91a1a}-1.47\%$
test_dqn_speed 3.9452ms 1.4445ms 692.2688 Ops/s 645.8949 Ops/s $\textbf{\color{#35bf28}+7.18\%}$
test_ddpg_speed 2.9466ms 2.7345ms 365.6909 Ops/s 361.1414 Ops/s $\color{#35bf28}+1.26\%$
test_sac_speed 9.0094ms 8.5061ms 117.5621 Ops/s 116.7804 Ops/s $\color{#35bf28}+0.67\%$
test_redq_speed 12.5932ms 10.6356ms 94.0236 Ops/s 93.7034 Ops/s $\color{#35bf28}+0.34\%$
test_redq_deprec_speed 12.5078ms 11.7827ms 84.8698 Ops/s 82.4138 Ops/s $\color{#35bf28}+2.98\%$
test_td3_speed 10.4524ms 8.6837ms 115.1584 Ops/s 115.0680 Ops/s $\color{#35bf28}+0.08\%$
test_cql_speed 28.1114ms 25.8344ms 38.7081 Ops/s 37.9941 Ops/s $\color{#35bf28}+1.88\%$
test_a2c_speed 7.2356ms 5.5297ms 180.8431 Ops/s 176.1199 Ops/s $\color{#35bf28}+2.68\%$
test_ppo_speed 6.0653ms 5.8237ms 171.7112 Ops/s 165.5896 Ops/s $\color{#35bf28}+3.70\%$
test_reinforce_speed 5.2643ms 4.5025ms 222.1013 Ops/s 213.9263 Ops/s $\color{#35bf28}+3.82\%$
test_iql_speed 20.4890ms 19.6539ms 50.8805 Ops/s 49.2180 Ops/s $\color{#35bf28}+3.38\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.8064ms 3.5540ms 281.3731 Ops/s 280.5757 Ops/s $\color{#35bf28}+0.28\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7316ms 0.5585ms 1.7904 KOps/s 1.7799 KOps/s $\color{#35bf28}+0.59\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7303ms 0.5320ms 1.8798 KOps/s 1.8793 KOps/s $\color{#35bf28}+0.03\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.8358ms 3.5771ms 279.5528 Ops/s 278.6175 Ops/s $\color{#35bf28}+0.34\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7197ms 0.5507ms 1.8157 KOps/s 1.7965 KOps/s $\color{#35bf28}+1.07\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7156ms 0.5251ms 1.9044 KOps/s 1.8804 KOps/s $\color{#35bf28}+1.28\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.8568ms 3.6813ms 271.6432 Ops/s 269.5669 Ops/s $\color{#35bf28}+0.77\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 91.7310ms 0.7653ms 1.3066 KOps/s 1.4497 KOps/s $\textbf{\color{#d91a1a}-9.87\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8462ms 0.6579ms 1.5200 KOps/s 1.5132 KOps/s $\color{#35bf28}+0.45\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.7088ms 3.5359ms 282.8106 Ops/s 282.5487 Ops/s $\color{#35bf28}+0.09\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8019ms 0.5599ms 1.7859 KOps/s 1.7582 KOps/s $\color{#35bf28}+1.58\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6991ms 0.5330ms 1.8761 KOps/s 1.8635 KOps/s $\color{#35bf28}+0.67\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.8639ms 3.5773ms 279.5439 Ops/s 280.6742 Ops/s $\color{#d91a1a}-0.40\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7512ms 0.5540ms 1.8051 KOps/s 1.7910 KOps/s $\color{#35bf28}+0.79\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7604ms 0.5272ms 1.8969 KOps/s 1.8727 KOps/s $\color{#35bf28}+1.29\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.8582ms 3.6888ms 271.0930 Ops/s 271.5179 Ops/s $\color{#d91a1a}-0.16\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8684ms 0.6854ms 1.4590 KOps/s 1.4477 KOps/s $\color{#35bf28}+0.78\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8605ms 0.6604ms 1.5142 KOps/s 1.5001 KOps/s $\color{#35bf28}+0.94\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1003s 11.0041ms 90.8752 Ops/s 102.4908 Ops/s $\textbf{\color{#d91a1a}-11.33\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 22.2253ms 15.8904ms 62.9309 Ops/s 62.6632 Ops/s $\color{#35bf28}+0.43\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.4731ms 3.0233ms 330.7617 Ops/s 325.3495 Ops/s $\color{#35bf28}+1.66\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 99.6734ms 10.9914ms 90.9799 Ops/s 88.8815 Ops/s $\color{#35bf28}+2.36\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 17.9838ms 15.7290ms 63.5767 Ops/s 62.3388 Ops/s $\color{#35bf28}+1.99\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.8129ms 3.0270ms 330.3582 Ops/s 326.7188 Ops/s $\color{#35bf28}+1.11\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 99.5977ms 11.3283ms 88.2743 Ops/s 86.1918 Ops/s $\color{#35bf28}+2.42\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 18.3134ms 15.9962ms 62.5147 Ops/s 61.4014 Ops/s $\color{#35bf28}+1.81\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.4467ms 3.3089ms 302.2109 Ops/s 296.6758 Ops/s $\color{#35bf28}+1.87\%$

@vmoens vmoens merged commit c45ee1f into main Feb 20, 2024
65 of 68 checks passed
@vmoens vmoens deleted the extend-with-list branch February 20, 2024 17:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Suitable for minor Suitable to be integrated in minor release (no new feature)
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[BUG] Unintended and surprising slicing of tensors with ReplayBuffer extend and sample.
2 participants