-
Notifications
You must be signed in to change notification settings - Fork 417
[Refactor] Limit the deepcopies in collectors #2451
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
Merged
Merged
Conversation
This file contains hidden or 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/2451
Note: Links to docs will display an error until the docs builds have been completed. ❌ 8 New Failures, 5 Unrelated FailuresAs of commit 383d9d8 with merge base b4d543e ( NEW FAILURES - The following jobs have failed:
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. |
|
| Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
|---|---|---|---|---|---|
| test_single | 60.1407ms | 59.3979ms | 16.8356 Ops/s | 16.5611 Ops/s | |
| test_sync | 45.0964ms | 32.8586ms | 30.4334 Ops/s | 30.9873 Ops/s | |
| test_async | 71.3078ms | 30.8663ms | 32.3978 Ops/s | 32.1109 Ops/s | |
| test_simple | 0.5002s | 0.4199s | 2.3817 Ops/s | 2.4207 Ops/s | |
| test_transformed | 0.6477s | 0.5783s | 1.7291 Ops/s | 1.6969 Ops/s | |
| test_serial | 1.3218s | 1.2597s | 0.7938 Ops/s | 0.7665 Ops/s | |
| test_parallel | 1.2149s | 1.1333s | 0.8824 Ops/s | 0.8892 Ops/s | |
| test_step_mdp_speed[True-True-True-True-True] | 0.2222ms | 28.0959μs | 35.5924 KOps/s | 36.1388 KOps/s | |
| test_step_mdp_speed[True-True-True-True-False] | 52.2780μs | 16.4321μs | 60.8564 KOps/s | 62.3339 KOps/s | |
| test_step_mdp_speed[True-True-True-False-True] | 0.5732ms | 15.9347μs | 62.7561 KOps/s | 62.9690 KOps/s | |
| test_step_mdp_speed[True-True-True-False-False] | 0.1046ms | 9.3392μs | 107.0752 KOps/s | 107.8260 KOps/s | |
| test_step_mdp_speed[True-True-False-True-True] | 70.8130μs | 29.8766μs | 33.4710 KOps/s | 33.9546 KOps/s | |
| test_step_mdp_speed[True-True-False-True-False] | 73.2160μs | 17.9605μs | 55.6778 KOps/s | 56.1432 KOps/s | |
| test_step_mdp_speed[True-True-False-False-True] | 49.4430μs | 17.6038μs | 56.8059 KOps/s | 56.9542 KOps/s | |
| test_step_mdp_speed[True-True-False-False-False] | 41.5280μs | 10.9313μs | 91.4807 KOps/s | 90.8220 KOps/s | |
| test_step_mdp_speed[True-False-True-True-True] | 83.2260μs | 31.4998μs | 31.7462 KOps/s | 32.1232 KOps/s | |
| test_step_mdp_speed[True-False-True-True-False] | 79.5490μs | 19.7083μs | 50.7401 KOps/s | 51.1539 KOps/s | |
| test_step_mdp_speed[True-False-True-False-True] | 55.0230μs | 17.7080μs | 56.4716 KOps/s | 56.4149 KOps/s | |
| test_step_mdp_speed[True-False-True-False-False] | 39.6140μs | 11.0338μs | 90.6306 KOps/s | 90.5063 KOps/s | |
| test_step_mdp_speed[True-False-False-True-True] | 85.5000μs | 32.8829μs | 30.4109 KOps/s | 30.6983 KOps/s | |
| test_step_mdp_speed[True-False-False-True-False] | 66.4050μs | 21.1795μs | 47.2155 KOps/s | 48.1945 KOps/s | |
| test_step_mdp_speed[True-False-False-False-True] | 54.9630μs | 19.2024μs | 52.0769 KOps/s | 53.2431 KOps/s | |
| test_step_mdp_speed[True-False-False-False-False] | 43.4710μs | 12.4936μs | 80.0409 KOps/s | 80.5660 KOps/s | |
| test_step_mdp_speed[False-True-True-True-True] | 82.9780μs | 31.5461μs | 31.6997 KOps/s | 32.2666 KOps/s | |
| test_step_mdp_speed[False-True-True-True-False] | 59.6720μs | 19.7588μs | 50.6103 KOps/s | 51.8235 KOps/s | |
| test_step_mdp_speed[False-True-True-False-True] | 51.9280μs | 20.1351μs | 49.6645 KOps/s | 50.3682 KOps/s | |
| test_step_mdp_speed[False-True-True-False-False] | 42.9810μs | 12.2164μs | 81.8569 KOps/s | 81.9693 KOps/s | |
| test_step_mdp_speed[False-True-False-True-True] | 0.6607ms | 33.2183μs | 30.1039 KOps/s | 31.0214 KOps/s | |
| test_step_mdp_speed[False-True-False-True-False] | 51.2760μs | 21.2321μs | 47.0985 KOps/s | 47.5832 KOps/s | |
| test_step_mdp_speed[False-True-False-False-True] | 3.1471ms | 21.6415μs | 46.2076 KOps/s | 44.8171 KOps/s | |
| test_step_mdp_speed[False-True-False-False-False] | 39.8350μs | 13.8336μs | 72.2878 KOps/s | 72.5836 KOps/s | |
| test_step_mdp_speed[False-False-True-True-True] | 85.7610μs | 35.1687μs | 28.4344 KOps/s | 29.2447 KOps/s | |
| test_step_mdp_speed[False-False-True-True-False] | 60.7840μs | 22.9578μs | 43.5581 KOps/s | 44.0835 KOps/s | |
| test_step_mdp_speed[False-False-True-False-True] | 70.3520μs | 21.5905μs | 46.3166 KOps/s | 46.8609 KOps/s | |
| test_step_mdp_speed[False-False-True-False-False] | 45.6250μs | 13.9426μs | 71.7228 KOps/s | 72.5338 KOps/s | |
| test_step_mdp_speed[False-False-False-True-True] | 91.0110μs | 35.9020μs | 27.8536 KOps/s | 28.3328 KOps/s | |
| test_step_mdp_speed[False-False-False-True-False] | 56.6160μs | 24.3666μs | 41.0398 KOps/s | 41.6382 KOps/s | |
| test_step_mdp_speed[False-False-False-False-True] | 97.6810μs | 22.8589μs | 43.7466 KOps/s | 43.8406 KOps/s | |
| test_step_mdp_speed[False-False-False-False-False] | 48.4420μs | 15.2091μs | 65.7502 KOps/s | 66.1598 KOps/s | |
| test_values[generalized_advantage_estimate-True-True] | 10.1798ms | 9.3794ms | 106.6169 Ops/s | 107.3091 Ops/s | |
| test_values[vec_generalized_advantage_estimate-True-True] | 41.0675ms | 33.7817ms | 29.6018 Ops/s | 27.7903 Ops/s | |
| test_values[td0_return_estimate-False-False] | 0.2707ms | 0.1777ms | 5.6271 KOps/s | 6.0245 KOps/s | |
| test_values[td1_return_estimate-False-False] | 30.5549ms | 23.2383ms | 43.0324 Ops/s | 43.3337 Ops/s | |
| test_values[vec_td1_return_estimate-False-False] | 34.5158ms | 33.4304ms | 29.9129 Ops/s | 28.1831 Ops/s | |
| test_values[td_lambda_return_estimate-True-False] | 34.3015ms | 33.6828ms | 29.6887 Ops/s | 29.8096 Ops/s | |
| test_values[vec_td_lambda_return_estimate-True-False] | 34.7073ms | 33.5236ms | 29.8297 Ops/s | 28.1248 Ops/s | |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 8.2407ms | 8.0927ms | 123.5680 Ops/s | 121.2782 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 3.3445ms | 2.5527ms | 391.7382 Ops/s | 517.8673 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.4766ms | 0.3496ms | 2.8602 KOps/s | 2.7987 KOps/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 41.2659ms | 39.1603ms | 25.5361 Ops/s | 21.4401 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.9793ms | 3.0153ms | 331.6424 Ops/s | 326.9369 Ops/s | |
| test_dqn_speed[False-None] | 5.7048ms | 1.3047ms | 766.4411 Ops/s | 753.0528 Ops/s | |
| test_dqn_speed[False-backward] | 1.9525ms | 1.7960ms | 556.7800 Ops/s | 559.8023 Ops/s | |
| test_dqn_speed[True-None] | 0.7082ms | 0.4488ms | 2.2280 KOps/s | 2.1747 KOps/s | |
| test_dqn_speed[True-backward] | 0.9217ms | 0.8634ms | 1.1582 KOps/s | 1.1189 KOps/s | |
| test_dqn_speed[reduce-overhead-None] | 1.6867ms | 0.4477ms | 2.2335 KOps/s | 2.1112 KOps/s | |
| test_dqn_speed[reduce-overhead-backward] | 0.9553ms | 0.8722ms | 1.1465 KOps/s | 1.1272 KOps/s | |
| test_ddpg_speed[False-None] | 4.1930ms | 2.7849ms | 359.0828 Ops/s | 359.5118 Ops/s | |
| test_ddpg_speed[False-backward] | 4.0632ms | 3.9289ms | 254.5229 Ops/s | 257.3439 Ops/s | |
| test_ddpg_speed[True-None] | 1.3874ms | 0.9796ms | 1.0208 KOps/s | 952.2182 Ops/s | |
| test_ddpg_speed[True-backward] | 2.2218ms | 1.8972ms | 527.0945 Ops/s | 512.8284 Ops/s | |
| test_ddpg_speed[reduce-overhead-None] | 1.3837ms | 0.9852ms | 1.0150 KOps/s | 975.4187 Ops/s | |
| test_ddpg_speed[reduce-overhead-backward] | 1.9714ms | 1.8736ms | 533.7380 Ops/s | 530.8524 Ops/s | |
| test_sac_speed[False-None] | 9.5908ms | 7.8095ms | 128.0489 Ops/s | 101.6244 Ops/s | |
| test_sac_speed[False-backward] | 11.1685ms | 10.6006ms | 94.3346 Ops/s | 94.3172 Ops/s | |
| test_sac_speed[True-None] | 2.0989ms | 1.8114ms | 552.0703 Ops/s | 538.2335 Ops/s | |
| test_sac_speed[True-backward] | 3.9149ms | 3.6502ms | 273.9547 Ops/s | 279.8854 Ops/s | |
| test_sac_speed[reduce-overhead-None] | 2.2587ms | 1.8335ms | 545.3934 Ops/s | 529.4534 Ops/s | |
| test_sac_speed[reduce-overhead-backward] | 4.0452ms | 3.6633ms | 272.9768 Ops/s | 282.1379 Ops/s | |
| test_redq_speed[False-None] | 14.7562ms | 13.0503ms | 76.6267 Ops/s | 76.9784 Ops/s | |
| test_redq_speed[False-backward] | 24.9807ms | 22.2356ms | 44.9728 Ops/s | 45.6069 Ops/s | |
| test_redq_speed[True-None] | 5.4117ms | 4.7177ms | 211.9678 Ops/s | 210.1638 Ops/s | |
| test_redq_speed[True-backward] | 13.0794ms | 12.3437ms | 81.0128 Ops/s | 81.8562 Ops/s | |
| test_redq_speed[reduce-overhead-None] | 5.1229ms | 4.5578ms | 219.4024 Ops/s | 198.6486 Ops/s | |
| test_redq_speed[reduce-overhead-backward] | 12.0659ms | 11.7231ms | 85.3016 Ops/s | 80.1918 Ops/s | |
| test_redq_deprec_speed[False-None] | 13.7163ms | 12.1330ms | 82.4198 Ops/s | 76.5608 Ops/s | |
| test_redq_deprec_speed[False-backward] | 20.3025ms | 17.5828ms | 56.8738 Ops/s | 52.3290 Ops/s | |
| test_redq_deprec_speed[True-None] | 3.8319ms | 3.5026ms | 285.5022 Ops/s | 278.1275 Ops/s | |
| test_redq_deprec_speed[True-backward] | 9.5978ms | 8.0678ms | 123.9501 Ops/s | 118.5249 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-None] | 4.4001ms | 3.5684ms | 280.2393 Ops/s | 262.7569 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-backward] | 9.2072ms | 8.2862ms | 120.6821 Ops/s | 123.8276 Ops/s | |
| test_td3_speed[False-None] | 8.2240ms | 7.6731ms | 130.3254 Ops/s | 127.6806 Ops/s | |
| test_td3_speed[False-backward] | 11.9878ms | 10.0273ms | 99.7277 Ops/s | 96.3864 Ops/s | |
| test_td3_speed[True-None] | 2.0530ms | 1.8846ms | 530.6075 Ops/s | 503.3320 Ops/s | |
| test_td3_speed[True-backward] | 4.0388ms | 3.7352ms | 267.7229 Ops/s | 274.2349 Ops/s | |
| test_td3_speed[reduce-overhead-None] | 2.1191ms | 1.9186ms | 521.2228 Ops/s | 505.8658 Ops/s | |
| test_td3_speed[reduce-overhead-backward] | 3.5926ms | 3.4696ms | 288.2148 Ops/s | 280.8458 Ops/s | |
| test_cql_speed[False-None] | 36.3689ms | 34.8957ms | 28.6568 Ops/s | 28.2067 Ops/s | |
| test_cql_speed[False-backward] | 48.1823ms | 44.8929ms | 22.2753 Ops/s | 22.2153 Ops/s | |
| test_cql_speed[True-None] | 16.1953ms | 15.4658ms | 64.6588 Ops/s | 63.3556 Ops/s | |
| test_cql_speed[True-backward] | 23.7590ms | 22.3733ms | 44.6961 Ops/s | 44.3086 Ops/s | |
| test_cql_speed[reduce-overhead-None] | 18.0187ms | 15.7621ms | 63.4435 Ops/s | 63.2177 Ops/s | |
| test_cql_speed[reduce-overhead-backward] | 23.4808ms | 22.1608ms | 45.1247 Ops/s | 45.0326 Ops/s | |
| test_a2c_speed[False-None] | 8.0267ms | 7.0099ms | 142.6546 Ops/s | 139.6111 Ops/s | |
| test_a2c_speed[False-backward] | 14.5386ms | 13.8972ms | 71.9571 Ops/s | 70.6309 Ops/s | |
| test_a2c_speed[True-None] | 3.8179ms | 3.2864ms | 304.2868 Ops/s | 300.3089 Ops/s | |
| test_a2c_speed[True-backward] | 10.4848ms | 9.8917ms | 101.0953 Ops/s | 102.5250 Ops/s | |
| test_a2c_speed[reduce-overhead-None] | 3.6225ms | 3.2926ms | 303.7119 Ops/s | 299.5496 Ops/s | |
| test_a2c_speed[reduce-overhead-backward] | 10.2404ms | 9.6661ms | 103.4548 Ops/s | 101.2136 Ops/s | |
| test_ppo_speed[False-None] | 8.6173ms | 7.3972ms | 135.1868 Ops/s | 133.6879 Ops/s | |
| test_ppo_speed[False-backward] | 15.2113ms | 14.5502ms | 68.7274 Ops/s | 69.2442 Ops/s | |
| test_ppo_speed[True-None] | 4.3887ms | 3.7781ms | 264.6844 Ops/s | 267.0438 Ops/s | |
| test_ppo_speed[True-backward] | 10.5724ms | 9.8115ms | 101.9208 Ops/s | 104.9604 Ops/s | |
| test_ppo_speed[reduce-overhead-None] | 4.4199ms | 3.8548ms | 259.4140 Ops/s | 266.7781 Ops/s | |
| test_ppo_speed[reduce-overhead-backward] | 11.1352ms | 10.1596ms | 98.4293 Ops/s | 102.7250 Ops/s | |
| test_reinforce_speed[False-None] | 8.2825ms | 6.9686ms | 143.5002 Ops/s | 154.6053 Ops/s | |
| test_reinforce_speed[False-backward] | 11.6698ms | 10.9685ms | 91.1700 Ops/s | 102.8298 Ops/s | |
| test_reinforce_speed[True-None] | 3.2382ms | 2.6597ms | 375.9857 Ops/s | 370.7941 Ops/s | |
| test_reinforce_speed[True-backward] | 9.2131ms | 8.5549ms | 116.8917 Ops/s | 114.5651 Ops/s | |
| test_reinforce_speed[reduce-overhead-None] | 3.1750ms | 2.6455ms | 378.0016 Ops/s | 374.7047 Ops/s | |
| test_reinforce_speed[reduce-overhead-backward] | 9.7444ms | 8.6382ms | 115.7645 Ops/s | 116.0784 Ops/s | |
| test_iql_speed[False-None] | 34.3517ms | 32.3436ms | 30.9180 Ops/s | 31.2053 Ops/s | |
| test_iql_speed[False-backward] | 47.4208ms | 44.9530ms | 22.2455 Ops/s | 22.3418 Ops/s | |
| test_iql_speed[True-None] | 14.5912ms | 13.6950ms | 73.0191 Ops/s | 74.8222 Ops/s | |
| test_iql_speed[True-backward] | 26.5028ms | 25.1325ms | 39.7890 Ops/s | 41.2711 Ops/s | |
| test_iql_speed[reduce-overhead-None] | 15.2772ms | 14.0135ms | 71.3595 Ops/s | 73.7657 Ops/s | |
| test_iql_speed[reduce-overhead-backward] | 27.6675ms | 25.4557ms | 39.2839 Ops/s | 39.5486 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 5.4191ms | 5.2062ms | 192.0798 Ops/s | 183.7287 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.0454ms | 0.4722ms | 2.1178 KOps/s | 2.0678 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7658ms | 0.4475ms | 2.2347 KOps/s | 2.1605 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.8872ms | 5.0296ms | 198.8213 Ops/s | 190.0377 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.9638ms | 0.4670ms | 2.1413 KOps/s | 830.8154 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6778ms | 0.4423ms | 2.2607 KOps/s | 2.2422 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 2.4499ms | 1.5548ms | 643.1731 Ops/s | 624.0745 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 2.2026ms | 1.5059ms | 664.0523 Ops/s | 648.2175 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 8.2335ms | 5.3066ms | 188.4429 Ops/s | 186.2688 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 3.1284ms | 0.6031ms | 1.6582 KOps/s | 1.6349 KOps/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 1.0747ms | 0.5914ms | 1.6909 KOps/s | 1.6724 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 5.9203ms | 5.1940ms | 192.5306 Ops/s | 191.3769 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.9264ms | 0.4776ms | 2.0939 KOps/s | 2.0460 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 7.6877ms | 0.4675ms | 2.1391 KOps/s | 2.2281 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 8.0899ms | 5.1149ms | 195.5071 Ops/s | 191.0608 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.3869ms | 0.4670ms | 2.1414 KOps/s | 2.1347 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6840ms | 0.4463ms | 2.2408 KOps/s | 2.1950 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 8.7911ms | 5.2883ms | 189.0983 Ops/s | 187.5583 Ops/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 3.1223ms | 0.6173ms | 1.6199 KOps/s | 1.6268 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7888ms | 0.5784ms | 1.7289 KOps/s | 1.6898 KOps/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.3835s | 11.7085ms | 85.4080 Ops/s | 228.8857 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 8.1776ms | 2.2422ms | 446.0004 Ops/s | 457.8094 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 2.3540ms | 1.2637ms | 791.3486 Ops/s | 847.8882 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 6.9383ms | 4.1664ms | 240.0175 Ops/s | 237.8527 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 4.9522ms | 1.9688ms | 507.9127 Ops/s | 470.1179 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 1.7587ms | 1.1911ms | 839.5788 Ops/s | 712.1887 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.3703s | 11.5556ms | 86.5383 Ops/s | 237.0786 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 7.3901ms | 2.3182ms | 431.3690 Ops/s | 427.5029 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.8779ms | 1.4381ms | 695.3760 Ops/s | 707.9499 Ops/s |
|
| Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
|---|---|---|---|---|---|
| test_single | 0.1033s | 0.1024s | 9.7662 Ops/s | 9.6502 Ops/s | |
| test_sync | 93.3064ms | 89.6012ms | 11.1606 Ops/s | 10.8126 Ops/s | |
| test_async | 0.2744s | 87.0641ms | 11.4858 Ops/s | 11.7099 Ops/s | |
| test_single_pixels | 0.1097s | 0.1093s | 9.1503 Ops/s | 9.1176 Ops/s | |
| test_sync_pixels | 84.5855ms | 73.1175ms | 13.6766 Ops/s | 13.8953 Ops/s | |
| test_async_pixels | 0.1370s | 68.0230ms | 14.7009 Ops/s | 14.6182 Ops/s | |
| test_simple | 0.7413s | 0.7400s | 1.3514 Ops/s | 1.3098 Ops/s | |
| test_transformed | 0.9708s | 0.9686s | 1.0324 Ops/s | 1.0124 Ops/s | |
| test_serial | 2.0965s | 2.0855s | 0.4795 Ops/s | 0.4744 Ops/s | |
| test_parallel | 1.8841s | 1.8659s | 0.5359 Ops/s | 0.5351 Ops/s | |
| test_step_mdp_speed[True-True-True-True-True] | 0.1824ms | 37.6895μs | 26.5326 KOps/s | 25.8546 KOps/s | |
| test_step_mdp_speed[True-True-True-True-False] | 0.1213ms | 21.2060μs | 47.1565 KOps/s | 46.4327 KOps/s | |
| test_step_mdp_speed[True-True-True-False-True] | 0.1612ms | 21.2337μs | 47.0950 KOps/s | 46.6351 KOps/s | |
| test_step_mdp_speed[True-True-True-False-False] | 0.1401ms | 12.0358μs | 83.0856 KOps/s | 81.6694 KOps/s | |
| test_step_mdp_speed[True-True-False-True-True] | 72.4130μs | 39.8675μs | 25.0831 KOps/s | 25.2654 KOps/s | |
| test_step_mdp_speed[True-True-False-True-False] | 0.1372ms | 23.6152μs | 42.3456 KOps/s | 42.0408 KOps/s | |
| test_step_mdp_speed[True-True-False-False-True] | 52.8920μs | 23.3074μs | 42.9048 KOps/s | 41.9730 KOps/s | |
| test_step_mdp_speed[True-True-False-False-False] | 56.8630μs | 14.1717μs | 70.5631 KOps/s | 70.1431 KOps/s | |
| test_step_mdp_speed[True-False-True-True-True] | 0.1982ms | 42.2208μs | 23.6850 KOps/s | 23.5930 KOps/s | |
| test_step_mdp_speed[True-False-True-True-False] | 0.1493ms | 25.4130μs | 39.3500 KOps/s | 38.9029 KOps/s | |
| test_step_mdp_speed[True-False-True-False-True] | 0.2361ms | 22.9766μs | 43.5226 KOps/s | 42.3730 KOps/s | |
| test_step_mdp_speed[True-False-True-False-False] | 44.0620μs | 14.0510μs | 71.1691 KOps/s | 69.5371 KOps/s | |
| test_step_mdp_speed[True-False-False-True-True] | 84.7530μs | 43.7142μs | 22.8759 KOps/s | 22.3540 KOps/s | |
| test_step_mdp_speed[True-False-False-True-False] | 54.2630μs | 27.2749μs | 36.6638 KOps/s | 36.1300 KOps/s | |
| test_step_mdp_speed[True-False-False-False-True] | 62.1730μs | 25.1304μs | 39.7925 KOps/s | 38.8141 KOps/s | |
| test_step_mdp_speed[True-False-False-False-False] | 45.9420μs | 16.0426μs | 62.3340 KOps/s | 61.4330 KOps/s | |
| test_step_mdp_speed[False-True-True-True-True] | 73.9530μs | 41.7375μs | 23.9593 KOps/s | 23.9747 KOps/s | |
| test_step_mdp_speed[False-True-True-True-False] | 55.2520μs | 25.0459μs | 39.9268 KOps/s | 38.9267 KOps/s | |
| test_step_mdp_speed[False-True-True-False-True] | 58.6720μs | 26.5733μs | 37.6318 KOps/s | 37.8223 KOps/s | |
| test_step_mdp_speed[False-True-True-False-False] | 57.6720μs | 15.5696μs | 64.2278 KOps/s | 63.0428 KOps/s | |
| test_step_mdp_speed[False-True-False-True-True] | 80.7530μs | 42.9595μs | 23.2777 KOps/s | 22.5690 KOps/s | |
| test_step_mdp_speed[False-True-False-True-False] | 0.1935ms | 27.2034μs | 36.7601 KOps/s | 35.7328 KOps/s | |
| test_step_mdp_speed[False-True-False-False-True] | 3.9881ms | 28.4013μs | 35.2096 KOps/s | 34.7260 KOps/s | |
| test_step_mdp_speed[False-True-False-False-False] | 62.1330μs | 17.8158μs | 56.1300 KOps/s | 55.6954 KOps/s | |
| test_step_mdp_speed[False-False-True-True-True] | 79.1330μs | 45.6464μs | 21.9075 KOps/s | 21.5486 KOps/s | |
| test_step_mdp_speed[False-False-True-True-False] | 79.3240μs | 29.1354μs | 34.3225 KOps/s | 33.2064 KOps/s | |
| test_step_mdp_speed[False-False-True-False-True] | 59.5530μs | 28.3739μs | 35.2437 KOps/s | 35.3719 KOps/s | |
| test_step_mdp_speed[False-False-True-False-False] | 50.7020μs | 17.4688μs | 57.2449 KOps/s | 55.5550 KOps/s | |
| test_step_mdp_speed[False-False-False-True-True] | 81.2330μs | 47.1825μs | 21.1943 KOps/s | 21.0000 KOps/s | |
| test_step_mdp_speed[False-False-False-True-False] | 73.7530μs | 31.1571μs | 32.0955 KOps/s | 30.8896 KOps/s | |
| test_step_mdp_speed[False-False-False-False-True] | 81.2240μs | 29.0641μs | 34.4067 KOps/s | 33.6817 KOps/s | |
| test_step_mdp_speed[False-False-False-False-False] | 67.6530μs | 19.2518μs | 51.9433 KOps/s | 50.2384 KOps/s | |
| test_values[generalized_advantage_estimate-True-True] | 26.6850ms | 25.5065ms | 39.2057 Ops/s | 39.1693 Ops/s | |
| test_values[vec_generalized_advantage_estimate-True-True] | 0.1022s | 2.9402ms | 340.1093 Ops/s | 333.2409 Ops/s | |
| test_values[td0_return_estimate-False-False] | 93.4840μs | 66.8350μs | 14.9622 KOps/s | 14.7358 KOps/s | |
| test_values[td1_return_estimate-False-False] | 61.1737ms | 58.7254ms | 17.0284 Ops/s | 17.4523 Ops/s | |
| test_values[vec_td1_return_estimate-False-False] | 1.4027ms | 1.0951ms | 913.1189 Ops/s | 924.9460 Ops/s | |
| test_values[td_lambda_return_estimate-True-False] | 96.6558ms | 93.5373ms | 10.6909 Ops/s | 11.0602 Ops/s | |
| test_values[vec_td_lambda_return_estimate-True-False] | 1.4662ms | 1.0789ms | 926.9086 Ops/s | 932.4140 Ops/s | |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 27.4317ms | 25.4470ms | 39.2973 Ops/s | 37.3807 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.0039ms | 0.7325ms | 1.3653 KOps/s | 1.3838 KOps/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.8491ms | 0.6690ms | 1.4948 KOps/s | 1.5092 KOps/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6228ms | 1.4734ms | 678.7021 Ops/s | 679.5196 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.8278ms | 0.6815ms | 1.4674 KOps/s | 1.4696 KOps/s | |
| test_dqn_speed[False-None] | 7.2349ms | 1.3624ms | 734.0218 Ops/s | 719.0175 Ops/s | |
| test_dqn_speed[False-backward] | 1.9966ms | 1.8583ms | 538.1187 Ops/s | 530.6309 Ops/s | |
| test_dqn_speed[True-None] | 0.7751ms | 0.5734ms | 1.7440 KOps/s | 1.6478 KOps/s | |
| test_dqn_speed[True-backward] | 1.1640ms | 1.0255ms | 975.1580 Ops/s | 971.9948 Ops/s | |
| test_dqn_speed[reduce-overhead-None] | 0.8582ms | 0.5685ms | 1.7590 KOps/s | 1.7282 KOps/s | |
| test_dqn_speed[reduce-overhead-backward] | 1.0434ms | 1.0007ms | 999.2996 Ops/s | 977.7280 Ops/s | |
| test_ddpg_speed[False-None] | 3.3222ms | 2.7027ms | 369.9935 Ops/s | 367.4625 Ops/s | |
| test_ddpg_speed[False-backward] | 4.3344ms | 3.9518ms | 253.0482 Ops/s | 251.5445 Ops/s | |
| test_ddpg_speed[True-None] | 1.6215ms | 1.2627ms | 791.9653 Ops/s | 790.1733 Ops/s | |
| test_ddpg_speed[True-backward] | 2.4182ms | 2.2385ms | 446.7261 Ops/s | 362.6855 Ops/s | |
| test_ddpg_speed[reduce-overhead-None] | 1.6105ms | 1.2663ms | 789.7302 Ops/s | 774.4218 Ops/s | |
| test_ddpg_speed[reduce-overhead-backward] | 2.4852ms | 2.2885ms | 436.9718 Ops/s | 445.5305 Ops/s | |
| test_sac_speed[False-None] | 8.8465ms | 7.5842ms | 131.8535 Ops/s | 132.7821 Ops/s | |
| test_sac_speed[False-backward] | 11.3370ms | 10.8292ms | 92.3429 Ops/s | 92.4952 Ops/s | |
| test_sac_speed[True-None] | 2.3704ms | 2.0541ms | 486.8396 Ops/s | 471.4595 Ops/s | |
| test_sac_speed[True-backward] | 4.1550ms | 3.9905ms | 250.5943 Ops/s | 233.6427 Ops/s | |
| test_sac_speed[reduce-overhead-None] | 2.4405ms | 2.0629ms | 484.7640 Ops/s | 490.4403 Ops/s | |
| test_sac_speed[reduce-overhead-backward] | 4.2124ms | 4.0338ms | 247.9034 Ops/s | 250.2101 Ops/s | |
| test_redq_speed[False-None] | 12.3255ms | 10.5180ms | 95.0751 Ops/s | 96.3215 Ops/s | |
| test_redq_speed[False-backward] | 19.0109ms | 17.9385ms | 55.7460 Ops/s | 55.5654 Ops/s | |
| test_redq_speed[True-None] | 3.8835ms | 3.4895ms | 286.5724 Ops/s | 282.2355 Ops/s | |
| test_redq_speed[True-backward] | 9.0006ms | 8.6098ms | 116.1473 Ops/s | 116.9056 Ops/s | |
| test_redq_speed[reduce-overhead-None] | 3.8965ms | 3.5497ms | 281.7100 Ops/s | 285.2855 Ops/s | |
| test_redq_speed[reduce-overhead-backward] | 9.2248ms | 8.6907ms | 115.0654 Ops/s | 116.0483 Ops/s | |
| test_redq_deprec_speed[False-None] | 12.4265ms | 10.5968ms | 94.3681 Ops/s | 91.5201 Ops/s | |
| test_redq_deprec_speed[False-backward] | 15.7941ms | 15.2510ms | 65.5693 Ops/s | 62.9027 Ops/s | |
| test_redq_deprec_speed[True-None] | 3.4621ms | 3.2429ms | 308.3706 Ops/s | 294.5042 Ops/s | |
| test_redq_deprec_speed[True-backward] | 7.4485ms | 7.0953ms | 140.9382 Ops/s | 136.3175 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-None] | 3.4847ms | 3.2247ms | 310.1089 Ops/s | 298.1876 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-backward] | 7.3826ms | 7.0559ms | 141.7260 Ops/s | 137.8068 Ops/s | |
| test_td3_speed[False-None] | 8.8853ms | 7.5722ms | 132.0618 Ops/s | 133.2994 Ops/s | |
| test_td3_speed[False-backward] | 11.0170ms | 10.4517ms | 95.6782 Ops/s | 95.6849 Ops/s | |
| test_td3_speed[True-None] | 2.1655ms | 2.0993ms | 476.3450 Ops/s | 474.9577 Ops/s | |
| test_td3_speed[True-backward] | 4.1236ms | 3.9413ms | 253.7216 Ops/s | 253.1056 Ops/s | |
| test_td3_speed[reduce-overhead-None] | 2.1691ms | 2.1205ms | 471.5813 Ops/s | 479.3887 Ops/s | |
| test_td3_speed[reduce-overhead-backward] | 4.0962ms | 3.9226ms | 254.9360 Ops/s | 253.5271 Ops/s | |
| test_cql_speed[False-None] | 32.0895ms | 26.2174ms | 38.1426 Ops/s | 39.8758 Ops/s | |
| test_cql_speed[False-backward] | 39.3866ms | 35.9985ms | 27.7790 Ops/s | 28.7822 Ops/s | |
| test_cql_speed[True-None] | 11.7231ms | 11.0277ms | 90.6808 Ops/s | 90.6759 Ops/s | |
| test_cql_speed[True-backward] | 17.3967ms | 17.0088ms | 58.7932 Ops/s | 56.6377 Ops/s | |
| test_cql_speed[reduce-overhead-None] | 11.7822ms | 11.0876ms | 90.1911 Ops/s | 91.5487 Ops/s | |
| test_cql_speed[reduce-overhead-backward] | 17.6697ms | 16.9939ms | 58.8445 Ops/s | 59.2713 Ops/s | |
| test_a2c_speed[False-None] | 5.7228ms | 5.3918ms | 185.4679 Ops/s | 183.3812 Ops/s | |
| test_a2c_speed[False-backward] | 12.2346ms | 11.8831ms | 84.1534 Ops/s | 83.5321 Ops/s | |
| test_a2c_speed[True-None] | 3.4466ms | 3.0819ms | 324.4803 Ops/s | 324.1413 Ops/s | |
| test_a2c_speed[True-backward] | 9.0668ms | 8.6721ms | 115.3127 Ops/s | 111.3970 Ops/s | |
| test_a2c_speed[reduce-overhead-None] | 3.4232ms | 3.0600ms | 326.7948 Ops/s | 320.9341 Ops/s | |
| test_a2c_speed[reduce-overhead-backward] | 9.0377ms | 8.6442ms | 115.6843 Ops/s | 117.0661 Ops/s | |
| test_ppo_speed[False-None] | 6.0089ms | 5.7102ms | 175.1248 Ops/s | 172.8717 Ops/s | |
| test_ppo_speed[False-backward] | 13.7733ms | 12.4914ms | 80.0551 Ops/s | 80.1431 Ops/s | |
| test_ppo_speed[True-None] | 3.8515ms | 3.4771ms | 287.5965 Ops/s | 284.8086 Ops/s | |
| test_ppo_speed[True-backward] | 8.9766ms | 8.4446ms | 118.4194 Ops/s | 119.6475 Ops/s | |
| test_ppo_speed[reduce-overhead-None] | 3.7113ms | 3.4572ms | 289.2546 Ops/s | 286.8805 Ops/s | |
| test_ppo_speed[reduce-overhead-backward] | 8.6308ms | 8.3932ms | 119.1442 Ops/s | 118.2386 Ops/s | |
| test_reinforce_speed[False-None] | 4.7841ms | 4.4970ms | 222.3724 Ops/s | 216.6402 Ops/s | |
| test_reinforce_speed[False-backward] | 7.9170ms | 7.4389ms | 134.4287 Ops/s | 132.6940 Ops/s | |
| test_reinforce_speed[True-None] | 2.5840ms | 2.2399ms | 446.4549 Ops/s | 442.8750 Ops/s | |
| test_reinforce_speed[True-backward] | 7.4702ms | 7.1579ms | 139.7067 Ops/s | 140.4228 Ops/s | |
| test_reinforce_speed[reduce-overhead-None] | 2.6466ms | 2.2534ms | 443.7753 Ops/s | 444.0050 Ops/s | |
| test_reinforce_speed[reduce-overhead-backward] | 7.3898ms | 7.1179ms | 140.4905 Ops/s | 139.9296 Ops/s | |
| test_iql_speed[False-None] | 20.7233ms | 19.6529ms | 50.8831 Ops/s | 50.3037 Ops/s | |
| test_iql_speed[False-backward] | 30.5176ms | 29.9738ms | 33.3624 Ops/s | 32.9429 Ops/s | |
| test_iql_speed[True-None] | 8.7821ms | 8.0326ms | 124.4923 Ops/s | 122.2189 Ops/s | |
| test_iql_speed[True-backward] | 17.5652ms | 16.8667ms | 59.2883 Ops/s | 60.5055 Ops/s | |
| test_iql_speed[reduce-overhead-None] | 8.4970ms | 8.0367ms | 124.4298 Ops/s | 124.1247 Ops/s | |
| test_iql_speed[reduce-overhead-backward] | 17.7972ms | 17.0124ms | 58.7806 Ops/s | 60.0844 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.0919ms | 6.8878ms | 145.1852 Ops/s | 144.9022 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.8097ms | 0.3401ms | 2.9403 KOps/s | 2.9430 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6291ms | 0.3263ms | 3.0642 KOps/s | 3.1299 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.1630ms | 6.7464ms | 148.2268 Ops/s | 148.5272 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.4522ms | 0.3357ms | 2.9786 KOps/s | 4.2051 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7369ms | 0.3217ms | 3.1086 KOps/s | 4.6287 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.7820ms | 1.4222ms | 703.1572 Ops/s | 815.5607 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5696ms | 1.3108ms | 762.8962 Ops/s | 795.4606 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 7.1168ms | 6.9145ms | 144.6229 Ops/s | 146.5918 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.6384ms | 0.4993ms | 2.0027 KOps/s | 2.1318 KOps/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7317ms | 0.4692ms | 2.1311 KOps/s | 2.4565 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.1213ms | 6.8485ms | 146.0164 Ops/s | 148.0702 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.1678ms | 0.3712ms | 2.6941 KOps/s | 4.0887 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6228ms | 0.3559ms | 2.8097 KOps/s | 4.5409 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.1775ms | 6.7387ms | 148.3972 Ops/s | 149.4406 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.9774ms | 0.3637ms | 2.7498 KOps/s | 3.2156 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6381ms | 0.3522ms | 2.8392 KOps/s | 3.4148 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 7.2097ms | 6.9312ms | 144.2748 Ops/s | 146.6318 Ops/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.1362ms | 0.5097ms | 1.9619 KOps/s | 2.2976 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7531ms | 0.4931ms | 2.0282 KOps/s | 2.7158 KOps/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4306s | 13.8583ms | 72.1588 Ops/s | 33.2815 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 3.7986ms | 1.5071ms | 663.5274 Ops/s | 694.2597 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 7.2799ms | 1.2334ms | 810.7552 Ops/s | 840.6918 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 6.9772ms | 5.3976ms | 185.2678 Ops/s | 176.8266 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 10.3840ms | 2.0738ms | 482.2158 Ops/s | 511.3577 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 2.2077ms | 1.1192ms | 893.5012 Ops/s | 942.4955 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.4316s | 14.1446ms | 70.6986 Ops/s | 175.4602 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 8.1737ms | 2.1164ms | 472.5055 Ops/s | 473.1449 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 8.7406ms | 1.4525ms | 688.4866 Ops/s | 711.3178 Ops/s |
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.
Refactoring
Refactoring of an existing 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.
Stack from ghstack (oldest at bottom):