-
Notifications
You must be signed in to change notification settings - Fork 325
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] Make sure ParallelEnv does not overflow mem when policy requires grad #1909
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1909
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (4 Unrelated Failures)As of commit a6dc5ee with merge base 899af07 (): FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
BROKEN TRUNK - The following jobs 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
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 15, 2024
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 62.7995ms | 61.4820ms | 16.2649 Ops/s | 15.8279 Ops/s | |
test_sync | 39.5130ms | 34.3400ms | 29.1206 Ops/s | 29.2080 Ops/s | |
test_async | 0.1132s | 33.3509ms | 29.9842 Ops/s | 30.0007 Ops/s | |
test_simple | 0.4912s | 0.4334s | 2.3071 Ops/s | 2.3069 Ops/s | |
test_transformed | 0.6501s | 0.5885s | 1.6992 Ops/s | 1.6601 Ops/s | |
test_serial | 1.4410s | 1.3870s | 0.7210 Ops/s | 0.6985 Ops/s | |
test_parallel | 1.4485s | 1.3635s | 0.7334 Ops/s | 0.7194 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1229ms | 22.0921μs | 45.2651 KOps/s | 46.7858 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 44.7030μs | 13.7366μs | 72.7984 KOps/s | 76.5652 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 69.3450μs | 12.9373μs | 77.2959 KOps/s | 81.2046 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 27.8120μs | 7.9096μs | 126.4293 KOps/s | 131.5163 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 71.6960μs | 23.2353μs | 43.0379 KOps/s | 44.2086 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 69.3790μs | 14.9107μs | 67.0661 KOps/s | 70.1783 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 47.9490μs | 14.1132μs | 70.8556 KOps/s | 73.5277 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 49.5920μs | 9.2167μs | 108.4986 KOps/s | 113.9253 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 96.2290μs | 24.6007μs | 40.6492 KOps/s | 42.1397 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 50.8440μs | 16.0838μs | 62.1744 KOps/s | 64.7744 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 60.8540μs | 14.0514μs | 71.1675 KOps/s | 73.3029 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 35.4750μs | 9.2752μs | 107.8144 KOps/s | 113.8768 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 54.5410μs | 25.7228μs | 38.8760 KOps/s | 39.4749 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 42.4290μs | 17.5763μs | 56.8948 KOps/s | 59.7091 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 41.3260μs | 15.2829μs | 65.4326 KOps/s | 68.1194 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 42.9700μs | 10.4068μs | 96.0906 KOps/s | 100.3458 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 61.8550μs | 24.6405μs | 40.5836 KOps/s | 41.8662 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 49.8820μs | 16.2882μs | 61.3942 KOps/s | 64.0533 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 43.8110μs | 16.5990μs | 60.2448 KOps/s | 62.7777 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 62.0750μs | 10.5242μs | 95.0193 KOps/s | 101.5099 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 49.9530μs | 26.2692μs | 38.0675 KOps/s | 39.5879 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 39.7240μs | 17.3803μs | 57.5362 KOps/s | 59.5516 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 58.2480μs | 17.6547μs | 56.6420 KOps/s | 58.7014 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 75.5140μs | 11.6376μs | 85.9282 KOps/s | 89.7495 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 74.7690μs | 27.0737μs | 36.9363 KOps/s | 37.8616 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 71.5600μs | 18.7578μs | 53.3112 KOps/s | 54.6296 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 39.5130μs | 17.5138μs | 57.0980 KOps/s | 58.4825 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 38.4710μs | 11.6634μs | 85.7380 KOps/s | 88.2413 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 72.3040μs | 27.9268μs | 35.8079 KOps/s | 36.4896 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 49.9830μs | 19.7985μs | 50.5090 KOps/s | 51.9299 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 70.3010μs | 18.5317μs | 53.9615 KOps/s | 55.0574 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 39.8450μs | 12.7183μs | 78.6268 KOps/s | 81.2215 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 9.8173ms | 9.3665ms | 106.7634 Ops/s | 106.9607 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 36.5676ms | 34.9805ms | 28.5874 Ops/s | 28.5927 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.2488ms | 0.1862ms | 5.3706 KOps/s | 6.1045 KOps/s | |
test_values[td1_return_estimate-False-False] | 24.2480ms | 23.4593ms | 42.6269 Ops/s | 42.9748 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 36.2597ms | 35.0198ms | 28.5553 Ops/s | 28.5991 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 37.1660ms | 33.9751ms | 29.4334 Ops/s | 29.8919 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 36.4439ms | 35.1062ms | 28.4850 Ops/s | 28.6252 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 8.7742ms | 8.1765ms | 122.3024 Ops/s | 123.5006 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.5843ms | 2.0178ms | 495.6011 Ops/s | 505.5291 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.4198ms | 0.3506ms | 2.8525 KOps/s | 2.8554 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 47.1950ms | 44.2662ms | 22.5906 Ops/s | 22.4911 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.6747ms | 3.0560ms | 327.2202 Ops/s | 329.8754 Ops/s | |
test_dqn_speed | 84.9814ms | 1.4822ms | 674.6808 Ops/s | 732.5672 Ops/s | |
test_ddpg_speed | 3.2131ms | 2.7066ms | 369.4704 Ops/s | 365.3198 Ops/s | |
test_sac_speed | 9.6430ms | 8.5465ms | 117.0065 Ops/s | 114.6919 Ops/s | |
test_redq_speed | 14.8184ms | 13.4319ms | 74.4496 Ops/s | 74.7168 Ops/s | |
test_redq_deprec_speed | 15.2998ms | 13.8768ms | 72.0626 Ops/s | 72.5528 Ops/s | |
test_td3_speed | 9.3515ms | 8.6912ms | 115.0594 Ops/s | 114.8332 Ops/s | |
test_cql_speed | 39.4712ms | 37.2444ms | 26.8497 Ops/s | 27.2565 Ops/s | |
test_a2c_speed | 8.7128ms | 7.4287ms | 134.6128 Ops/s | 136.1814 Ops/s | |
test_ppo_speed | 8.1016ms | 7.7143ms | 129.6295 Ops/s | 130.0405 Ops/s | |
test_reinforce_speed | 7.6342ms | 6.6915ms | 149.4426 Ops/s | 151.0444 Ops/s | |
test_iql_speed | 34.2456ms | 33.4238ms | 29.9188 Ops/s | 30.3905 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.2675ms | 2.8712ms | 348.2844 Ops/s | 371.0268 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7482ms | 0.5151ms | 1.9415 KOps/s | 1.9315 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.9524ms | 0.4942ms | 2.0237 KOps/s | 2.0183 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 4.0792ms | 2.8373ms | 352.4420 Ops/s | 358.2492 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7298ms | 0.5054ms | 1.9784 KOps/s | 1.9577 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.8015ms | 0.4848ms | 2.0625 KOps/s | 2.0213 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.8742ms | 3.0171ms | 331.4438 Ops/s | 353.1929 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9537ms | 0.6376ms | 1.5684 KOps/s | 1.5661 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8676ms | 0.6085ms | 1.6434 KOps/s | 1.6381 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.0556ms | 2.8194ms | 354.6880 Ops/s | 366.1384 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.9372ms | 0.5115ms | 1.9550 KOps/s | 1.9388 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7935ms | 0.4973ms | 2.0108 KOps/s | 2.0061 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.3149ms | 2.8464ms | 351.3206 Ops/s | 361.2859 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6591ms | 0.5066ms | 1.9738 KOps/s | 1.9443 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.8796ms | 0.4823ms | 2.0735 KOps/s | 2.0574 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 4.2413ms | 2.9166ms | 342.8699 Ops/s | 346.4617 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.7657ms | 0.6329ms | 1.5801 KOps/s | 1.4825 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8842ms | 0.6083ms | 1.6439 KOps/s | 1.6124 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 97.2091ms | 9.6169ms | 103.9837 Ops/s | 128.4099 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 15.8766ms | 13.4454ms | 74.3747 Ops/s | 75.5098 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 4.2742ms | 2.5296ms | 395.3126 Ops/s | 389.2594 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 98.8180ms | 7.8372ms | 127.5966 Ops/s | 101.7218 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 0.1089s | 15.3476ms | 65.1567 Ops/s | 75.9572 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 4.4154ms | 2.5563ms | 391.1906 Ops/s | 390.2340 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1010s | 8.1973ms | 121.9918 Ops/s | 100.2805 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 16.2835ms | 13.9803ms | 71.5294 Ops/s | 74.5045 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.1354ms | 2.7910ms | 358.2999 Ops/s | 363.3867 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 0.1171s | 0.1166s | 8.5790 Ops/s | 8.7313 Ops/s | |
test_sync | 0.1726s | 0.1033s | 9.6817 Ops/s | 9.6864 Ops/s | |
test_async | 0.1814s | 91.8909ms | 10.8825 Ops/s | 10.9853 Ops/s | |
test_single_pixels | 0.1372s | 0.1369s | 7.3049 Ops/s | 7.7975 Ops/s | |
test_sync_pixels | 83.0982ms | 81.6980ms | 12.2402 Ops/s | 12.6030 Ops/s | |
test_async_pixels | 0.1973s | 73.8068ms | 13.5489 Ops/s | 13.5869 Ops/s | |
test_simple | 0.9017s | 0.8312s | 1.2030 Ops/s | 1.2530 Ops/s | |
test_transformed | 1.1457s | 1.0783s | 0.9274 Ops/s | 0.9473 Ops/s | |
test_serial | 2.4850s | 2.4173s | 0.4137 Ops/s | 0.4305 Ops/s | |
test_parallel | 2.2438s | 2.1003s | 0.4761 Ops/s | 0.4849 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1087ms | 33.1781μs | 30.1404 KOps/s | 31.6514 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 45.1200μs | 20.0778μs | 49.8062 KOps/s | 52.0555 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 44.7400μs | 18.8198μs | 53.1355 KOps/s | 54.8190 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 29.5800μs | 11.0963μs | 90.1197 KOps/s | 90.8613 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 65.1510μs | 35.1316μs | 28.4644 KOps/s | 29.4216 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 43.9800μs | 21.8550μs | 45.7560 KOps/s | 48.1179 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 52.2100μs | 20.6961μs | 48.3184 KOps/s | 49.6362 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 45.1400μs | 13.0495μs | 76.6315 KOps/s | 78.1083 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 66.6910μs | 36.8048μs | 27.1704 KOps/s | 27.6057 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 38.0700μs | 23.6619μs | 42.2621 KOps/s | 42.9265 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 43.3610μs | 20.2099μs | 49.4807 KOps/s | 49.7479 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 33.9600μs | 13.2426μs | 75.5139 KOps/s | 77.6545 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 67.8700μs | 38.9478μs | 25.6754 KOps/s | 26.2797 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 48.8000μs | 25.4825μs | 39.2426 KOps/s | 39.8893 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 44.7300μs | 21.8639μs | 45.7376 KOps/s | 45.4866 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 31.0000μs | 14.9287μs | 66.9851 KOps/s | 68.5984 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 61.2200μs | 37.4921μs | 26.6723 KOps/s | 27.2273 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 45.5110μs | 23.4904μs | 42.5706 KOps/s | 42.8097 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 50.7600μs | 24.4130μs | 40.9618 KOps/s | 41.5490 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 41.4110μs | 15.1756μs | 65.8953 KOps/s | 68.3145 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 74.9610μs | 38.2747μs | 26.1269 KOps/s | 26.2526 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 48.8500μs | 25.8778μs | 38.6432 KOps/s | 39.5310 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 46.8300μs | 26.4293μs | 37.8368 KOps/s | 38.4507 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 38.6290μs | 16.9721μs | 58.9203 KOps/s | 60.3212 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 72.1010μs | 40.8893μs | 24.4563 KOps/s | 25.0030 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 49.4300μs | 27.6709μs | 36.1391 KOps/s | 36.6937 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 48.6100μs | 25.5553μs | 39.1309 KOps/s | 38.8748 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 40.3510μs | 16.6479μs | 60.0676 KOps/s | 60.5300 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 71.7810μs | 41.6364μs | 24.0175 KOps/s | 24.2425 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 56.0900μs | 29.3063μs | 34.1223 KOps/s | 34.6779 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 54.3400μs | 26.7308μs | 37.4101 KOps/s | 37.1442 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 40.1600μs | 18.3888μs | 54.3808 KOps/s | 55.0246 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 25.8157ms | 25.2773ms | 39.5612 Ops/s | 40.3367 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 84.9018ms | 3.2666ms | 306.1264 Ops/s | 297.1593 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.1005ms | 62.4134μs | 16.0222 KOps/s | 16.3744 KOps/s | |
test_values[td1_return_estimate-False-False] | 58.0219ms | 56.8943ms | 17.5764 Ops/s | 19.1727 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 2.1456ms | 1.8030ms | 554.6284 Ops/s | 569.7081 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 93.0446ms | 92.4699ms | 10.8143 Ops/s | 12.0133 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 3.8526ms | 1.8048ms | 554.0852 Ops/s | 558.0733 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 25.8219ms | 25.2145ms | 39.6596 Ops/s | 43.0827 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 0.8882ms | 0.7397ms | 1.3519 KOps/s | 1.4317 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7149ms | 0.6600ms | 1.5151 KOps/s | 1.5435 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.5355ms | 1.4649ms | 682.6274 Ops/s | 689.8699 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.9526ms | 0.6778ms | 1.4754 KOps/s | 1.5019 KOps/s | |
test_dqn_speed | 7.6218ms | 1.4542ms | 687.6615 Ops/s | 709.8561 Ops/s | |
test_ddpg_speed | 2.9403ms | 2.7984ms | 357.3429 Ops/s | 366.6892 Ops/s | |
test_sac_speed | 9.2945ms | 8.5025ms | 117.6121 Ops/s | 121.2321 Ops/s | |
test_redq_speed | 11.2397ms | 10.4562ms | 95.6374 Ops/s | 95.9347 Ops/s | |
test_redq_deprec_speed | 11.7906ms | 11.3444ms | 88.1489 Ops/s | 86.4788 Ops/s | |
test_td3_speed | 8.7752ms | 8.5569ms | 116.8648 Ops/s | 117.8800 Ops/s | |
test_cql_speed | 26.1641ms | 25.3395ms | 39.4641 Ops/s | 39.1418 Ops/s | |
test_a2c_speed | 96.3064ms | 5.9065ms | 169.3058 Ops/s | 188.7909 Ops/s | |
test_ppo_speed | 5.9689ms | 5.6793ms | 176.0772 Ops/s | 178.4285 Ops/s | |
test_reinforce_speed | 4.7286ms | 4.4667ms | 223.8812 Ops/s | 224.1307 Ops/s | |
test_iql_speed | 20.3487ms | 19.5888ms | 51.0497 Ops/s | 50.5973 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.8642ms | 3.6070ms | 277.2396 Ops/s | 284.3004 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7249ms | 0.5645ms | 1.7716 KOps/s | 1.7956 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6725ms | 0.5341ms | 1.8722 KOps/s | 1.8937 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.8791ms | 3.6223ms | 276.0676 Ops/s | 281.0566 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6986ms | 0.5533ms | 1.8074 KOps/s | 1.8251 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7070ms | 0.5270ms | 1.8975 KOps/s | 1.9264 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.8393ms | 3.7058ms | 269.8497 Ops/s | 273.1752 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.8208ms | 0.6821ms | 1.4661 KOps/s | 1.4757 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7811ms | 0.6547ms | 1.5274 KOps/s | 1.3591 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.7324ms | 3.5949ms | 278.1736 Ops/s | 281.1753 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6928ms | 0.5601ms | 1.7854 KOps/s | 1.7996 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6475ms | 0.5338ms | 1.8734 KOps/s | 1.6241 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.7836ms | 3.6127ms | 276.7989 Ops/s | 278.4871 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6853ms | 0.5530ms | 1.8084 KOps/s | 1.8239 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.8475ms | 0.5294ms | 1.8889 KOps/s | 1.9093 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.8331ms | 3.7386ms | 267.4806 Ops/s | 271.8693 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9131ms | 0.6828ms | 1.4646 KOps/s | 1.4799 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7988ms | 0.6616ms | 1.5115 KOps/s | 1.5354 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1242s | 10.1556ms | 98.4677 Ops/s | 81.5797 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 19.3582ms | 16.7313ms | 59.7681 Ops/s | 63.5015 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 8.0115ms | 3.1132ms | 321.2125 Ops/s | 335.7060 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.1187s | 12.2197ms | 81.8350 Ops/s | 100.5829 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 18.6853ms | 16.4032ms | 60.9637 Ops/s | 63.8073 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 8.0192ms | 3.0970ms | 322.8921 Ops/s | 333.5678 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1192s | 10.2588ms | 97.4775 Ops/s | 96.0935 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 19.5362ms | 16.8432ms | 59.3712 Ops/s | 62.6563 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 8.5091ms | 3.4337ms | 291.2339 Ops/s | 304.9046 Ops/s |
vmoens
added
bug
Something isn't working
Suitable for minor
Suitable to be integrated in minor release (no new feature)
labels
Feb 15, 2024
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)
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Also adds a test to make sure SerialEnv gradients work ok
cc @AndreaTirinzoni