-
Notifications
You must be signed in to change notification settings - Fork 350
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] TrajCounter transform #2532
Merged
Merged
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/2532
Note: Links to docs will display an error until the docs builds have been completed. ❌ 19 New Failures, 3 Unrelated FailuresAs of commit 7c41858 with merge base edbf3de ( 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. |
vmoens
added a commit
that referenced
this pull request
Nov 2, 2024
ghstack-source-id: 324e88a9cca8b31dc9f79de5c17c7287e711f154 Pull Request resolved: #2532
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_simple | 0.4189s | 0.4179s | 2.3927 Ops/s | 2.3057 Ops/s | |
test_transformed | 0.5915s | 0.5861s | 1.7061 Ops/s | 1.7393 Ops/s | |
test_serial | 1.3642s | 1.3271s | 0.7535 Ops/s | 0.7652 Ops/s | |
test_parallel | 1.3945s | 1.3137s | 0.7612 Ops/s | 0.7552 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.2181ms | 27.4958μs | 36.3692 KOps/s | 37.1206 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 43.9820μs | 15.8194μs | 63.2135 KOps/s | 63.2546 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 0.2744ms | 15.6928μs | 63.7236 KOps/s | 60.3842 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 59.7420μs | 8.9905μs | 111.2288 KOps/s | 111.4943 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 71.7940μs | 28.7418μs | 34.7925 KOps/s | 35.0029 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 51.6070μs | 17.4111μs | 57.4348 KOps/s | 57.1354 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 59.2710μs | 17.0452μs | 58.6676 KOps/s | 58.5112 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 41.0370μs | 10.6210μs | 94.1528 KOps/s | 93.7040 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 72.1950μs | 30.3691μs | 32.9282 KOps/s | 32.7646 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 54.5220μs | 19.0302μs | 52.5481 KOps/s | 52.2822 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 50.0430μs | 17.1413μs | 58.3385 KOps/s | 59.1776 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 44.9440μs | 10.5449μs | 94.8327 KOps/s | 94.9879 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 72.0250μs | 31.9718μs | 31.2776 KOps/s | 31.6266 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 94.4070μs | 20.4781μs | 48.8325 KOps/s | 48.0528 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 44.6040μs | 18.4869μs | 54.0923 KOps/s | 53.8678 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 35.4370μs | 12.2926μs | 81.3500 KOps/s | 83.3414 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 0.1536ms | 30.5564μs | 32.7263 KOps/s | 33.2507 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 50.6050μs | 18.9480μs | 52.7760 KOps/s | 52.0419 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 49.7630μs | 19.5067μs | 51.2646 KOps/s | 51.2464 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 36.0870μs | 11.8207μs | 84.5973 KOps/s | 84.8363 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 70.0410μs | 32.0857μs | 31.1665 KOps/s | 31.1666 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 58.7600μs | 20.8012μs | 48.0741 KOps/s | 48.2451 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 3.2936ms | 21.3946μs | 46.7407 KOps/s | 47.3743 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 40.7970μs | 13.5498μs | 73.8017 KOps/s | 74.2710 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 74.7690μs | 33.6844μs | 29.6873 KOps/s | 29.9166 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 48.2400μs | 22.0878μs | 45.2738 KOps/s | 44.9671 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 54.5820μs | 21.0742μs | 47.4515 KOps/s | 47.4193 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 44.1030μs | 13.5080μs | 74.0304 KOps/s | 74.0607 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 68.2180μs | 34.7469μs | 28.7796 KOps/s | 29.0440 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 57.4080μs | 23.6643μs | 42.2577 KOps/s | 42.0967 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 50.2040μs | 22.2559μs | 44.9320 KOps/s | 44.5029 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 41.8990μs | 14.9167μs | 67.0390 KOps/s | 66.9145 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 10.3141ms | 9.8922ms | 101.0896 Ops/s | 101.4725 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 37.5073ms | 35.5818ms | 28.1042 Ops/s | 27.5888 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.2022ms | 0.1797ms | 5.5661 KOps/s | 4.8594 KOps/s | |
test_values[td1_return_estimate-False-False] | 36.1122ms | 25.1918ms | 39.6955 Ops/s | 40.8135 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 39.0161ms | 35.8100ms | 27.9251 Ops/s | 27.2534 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 39.7758ms | 36.1302ms | 27.6777 Ops/s | 28.8992 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 53.5939ms | 36.5412ms | 27.3664 Ops/s | 27.5281 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 12.1385ms | 8.5999ms | 116.2798 Ops/s | 119.1723 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.2904ms | 1.9980ms | 500.5012 Ops/s | 501.7762 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.4847ms | 0.3606ms | 2.7730 KOps/s | 2.8016 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 49.4837ms | 46.8559ms | 21.3420 Ops/s | 21.2343 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.8829ms | 3.0426ms | 328.6671 Ops/s | 326.0905 Ops/s | |
test_dqn_speed[False-None] | 6.4138ms | 1.3644ms | 732.9317 Ops/s | 768.2091 Ops/s | |
test_dqn_speed[False-backward] | 1.9095ms | 1.8385ms | 543.9227 Ops/s | 557.3538 Ops/s | |
test_dqn_speed[True-None] | 1.4435ms | 0.4855ms | 2.0598 KOps/s | 2.1341 KOps/s | |
test_dqn_speed[True-backward] | 0.9442ms | 0.8996ms | 1.1117 KOps/s | 1.0839 KOps/s | |
test_dqn_speed[reduce-overhead-None] | 0.5913ms | 0.4787ms | 2.0888 KOps/s | 2.1213 KOps/s | |
test_dqn_speed[reduce-overhead-backward] | 1.0289ms | 0.9210ms | 1.0858 KOps/s | 1.1018 KOps/s | |
test_ddpg_speed[False-None] | 4.0793ms | 2.8411ms | 351.9717 Ops/s | 358.3360 Ops/s | |
test_ddpg_speed[False-backward] | 4.1540ms | 4.0188ms | 248.8323 Ops/s | 255.3215 Ops/s | |
test_ddpg_speed[True-None] | 1.5073ms | 1.0327ms | 968.2893 Ops/s | 989.6066 Ops/s | |
test_ddpg_speed[True-backward] | 2.0094ms | 1.9492ms | 513.0262 Ops/s | 446.6737 Ops/s | |
test_ddpg_speed[reduce-overhead-None] | 1.6419ms | 1.1064ms | 903.8311 Ops/s | 981.7677 Ops/s | |
test_ddpg_speed[reduce-overhead-backward] | 2.1014ms | 1.9740ms | 506.5731 Ops/s | 500.5337 Ops/s | |
test_sac_speed[False-None] | 0.2589s | 10.5248ms | 95.0141 Ops/s | 127.4975 Ops/s | |
test_sac_speed[False-backward] | 12.4302ms | 11.4943ms | 86.9996 Ops/s | 94.8859 Ops/s | |
test_sac_speed[True-None] | 3.5221ms | 1.9608ms | 509.9978 Ops/s | 540.7913 Ops/s | |
test_sac_speed[True-backward] | 4.1852ms | 3.9928ms | 250.4515 Ops/s | 265.7531 Ops/s | |
test_sac_speed[reduce-overhead-None] | 2.2023ms | 1.9036ms | 525.3289 Ops/s | 538.4732 Ops/s | |
test_sac_speed[reduce-overhead-backward] | 5.5338ms | 4.0127ms | 249.2110 Ops/s | 281.2990 Ops/s | |
test_redq_speed[False-None] | 15.0476ms | 13.7486ms | 72.7345 Ops/s | 72.2366 Ops/s | |
test_redq_speed[False-backward] | 24.8680ms | 23.3857ms | 42.7612 Ops/s | 44.1733 Ops/s | |
test_redq_speed[True-None] | 6.6458ms | 5.5041ms | 181.6827 Ops/s | 202.2777 Ops/s | |
test_redq_speed[True-backward] | 14.5925ms | 13.2183ms | 75.6525 Ops/s | 79.4821 Ops/s | |
test_redq_speed[reduce-overhead-None] | 6.2228ms | 5.5210ms | 181.1268 Ops/s | 217.6467 Ops/s | |
test_redq_speed[reduce-overhead-backward] | 15.0812ms | 13.1093ms | 76.2815 Ops/s | 80.4935 Ops/s | |
test_redq_deprec_speed[False-None] | 15.7846ms | 13.7419ms | 72.7700 Ops/s | 78.6444 Ops/s | |
test_redq_deprec_speed[False-backward] | 22.3549ms | 19.9467ms | 50.1337 Ops/s | 53.1402 Ops/s | |
test_redq_deprec_speed[True-None] | 4.0125ms | 3.6296ms | 275.5099 Ops/s | 277.7549 Ops/s | |
test_redq_deprec_speed[True-backward] | 9.0877ms | 8.0719ms | 123.8862 Ops/s | 122.0103 Ops/s | |
test_redq_deprec_speed[reduce-overhead-None] | 4.1357ms | 3.6526ms | 273.7783 Ops/s | 276.7791 Ops/s | |
test_redq_deprec_speed[reduce-overhead-backward] | 8.5348ms | 8.0259ms | 124.5963 Ops/s | 124.6535 Ops/s | |
test_td3_speed[False-None] | 10.3652ms | 7.7810ms | 128.5188 Ops/s | 129.4875 Ops/s | |
test_td3_speed[False-backward] | 10.6080ms | 10.1720ms | 98.3092 Ops/s | 98.3895 Ops/s | |
test_td3_speed[True-None] | 1.9744ms | 1.7434ms | 573.5972 Ops/s | 571.5312 Ops/s | |
test_td3_speed[True-backward] | 5.3367ms | 3.4895ms | 286.5722 Ops/s | 298.5615 Ops/s | |
test_td3_speed[reduce-overhead-None] | 2.0400ms | 1.7296ms | 578.1756 Ops/s | 570.0942 Ops/s | |
test_td3_speed[reduce-overhead-backward] | 3.3713ms | 3.3191ms | 301.2842 Ops/s | 295.1733 Ops/s | |
test_cql_speed[False-None] | 41.9823ms | 35.9747ms | 27.7973 Ops/s | 27.9749 Ops/s | |
test_cql_speed[False-backward] | 59.9068ms | 47.4357ms | 21.0812 Ops/s | 21.7017 Ops/s | |
test_cql_speed[True-None] | 16.6732ms | 15.3226ms | 65.2630 Ops/s | 64.2032 Ops/s | |
test_cql_speed[True-backward] | 23.0097ms | 21.8603ms | 45.7451 Ops/s | 44.1639 Ops/s | |
test_cql_speed[reduce-overhead-None] | 16.1565ms | 15.3189ms | 65.2790 Ops/s | 64.1246 Ops/s | |
test_cql_speed[reduce-overhead-backward] | 22.2986ms | 21.8146ms | 45.8408 Ops/s | 44.4230 Ops/s | |
test_a2c_speed[False-None] | 8.1429ms | 7.0635ms | 141.5730 Ops/s | 140.0833 Ops/s | |
test_a2c_speed[False-backward] | 15.3921ms | 14.0847ms | 70.9991 Ops/s | 70.0162 Ops/s | |
test_a2c_speed[True-None] | 3.8840ms | 3.3005ms | 302.9832 Ops/s | 297.1048 Ops/s | |
test_a2c_speed[True-backward] | 10.0449ms | 9.7019ms | 103.0730 Ops/s | 102.4934 Ops/s | |
test_a2c_speed[reduce-overhead-None] | 3.6720ms | 3.3135ms | 301.7989 Ops/s | 297.6720 Ops/s | |
test_a2c_speed[reduce-overhead-backward] | 9.9521ms | 9.6439ms | 103.6924 Ops/s | 102.6753 Ops/s | |
test_ppo_speed[False-None] | 7.9372ms | 7.3225ms | 136.5656 Ops/s | 136.1124 Ops/s | |
test_ppo_speed[False-backward] | 16.2112ms | 14.3972ms | 69.4579 Ops/s | 68.7837 Ops/s | |
test_ppo_speed[True-None] | 4.1179ms | 3.7009ms | 270.2067 Ops/s | 266.4545 Ops/s | |
test_ppo_speed[True-backward] | 10.3148ms | 9.5633ms | 104.5669 Ops/s | 104.0329 Ops/s | |
test_ppo_speed[reduce-overhead-None] | 4.2848ms | 3.6897ms | 271.0215 Ops/s | 269.0026 Ops/s | |
test_ppo_speed[reduce-overhead-backward] | 9.8800ms | 9.5401ms | 104.8208 Ops/s | 103.6449 Ops/s | |
test_reinforce_speed[False-None] | 7.3681ms | 6.4261ms | 155.6148 Ops/s | 155.1062 Ops/s | |
test_reinforce_speed[False-backward] | 10.4918ms | 9.6955ms | 103.1408 Ops/s | 103.0505 Ops/s | |
test_reinforce_speed[True-None] | 2.9657ms | 2.6336ms | 379.7151 Ops/s | 371.7326 Ops/s | |
test_reinforce_speed[True-backward] | 9.5237ms | 8.8282ms | 113.2731 Ops/s | 114.6976 Ops/s | |
test_reinforce_speed[reduce-overhead-None] | 2.9579ms | 2.6432ms | 378.3279 Ops/s | 375.4496 Ops/s | |
test_reinforce_speed[reduce-overhead-backward] | 8.8390ms | 8.4966ms | 117.6936 Ops/s | 116.0822 Ops/s | |
test_iql_speed[False-None] | 33.7543ms | 31.3088ms | 31.9399 Ops/s | 31.2551 Ops/s | |
test_iql_speed[False-backward] | 46.7028ms | 44.2124ms | 22.6181 Ops/s | 21.4515 Ops/s | |
test_iql_speed[True-None] | 11.4736ms | 10.5135ms | 95.1155 Ops/s | 87.9104 Ops/s | |
test_iql_speed[True-backward] | 21.9756ms | 21.2145ms | 47.1376 Ops/s | 44.8834 Ops/s | |
test_iql_speed[reduce-overhead-None] | 11.4360ms | 10.6727ms | 93.6970 Ops/s | 88.5045 Ops/s | |
test_iql_speed[reduce-overhead-backward] | 22.1260ms | 21.2976ms | 46.9537 Ops/s | 44.2741 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.4049ms | 4.7503ms | 210.5127 Ops/s | 194.5210 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.9373ms | 0.5024ms | 1.9905 KOps/s | 1.9246 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6512ms | 0.4702ms | 2.1268 KOps/s | 2.0378 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 11.6857ms | 4.7022ms | 212.6664 Ops/s | 203.0198 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7450ms | 0.4869ms | 2.0537 KOps/s | 1.9955 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.8325ms | 0.4669ms | 2.1416 KOps/s | 2.0730 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 2.3778ms | 1.6787ms | 595.7091 Ops/s | 606.3569 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 2.7633ms | 1.5835ms | 631.5210 Ops/s | 621.9022 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.0983ms | 4.6765ms | 213.8349 Ops/s | 196.7453 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.2106ms | 0.6339ms | 1.5776 KOps/s | 1.5380 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.9956ms | 0.6110ms | 1.6368 KOps/s | 1.5869 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.1660ms | 4.5516ms | 219.7006 Ops/s | 202.9627 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.1833ms | 0.5046ms | 1.9816 KOps/s | 1.9412 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6396ms | 0.4737ms | 2.1109 KOps/s | 2.0107 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.3610ms | 4.5042ms | 222.0155 Ops/s | 205.0377 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7803ms | 0.4925ms | 2.0305 KOps/s | 2.0114 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 8.0968ms | 0.4758ms | 2.1018 KOps/s | 2.0521 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.7628ms | 4.6677ms | 214.2390 Ops/s | 200.6359 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.0973ms | 0.6377ms | 1.5681 KOps/s | 1.5072 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.9703ms | 0.6120ms | 1.6339 KOps/s | 1.5946 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 5.3957ms | 4.1511ms | 240.8983 Ops/s | 228.9607 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 4.9153ms | 2.2802ms | 438.5625 Ops/s | 393.2100 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 7.1696ms | 1.3758ms | 726.8706 Ops/s | 611.7705 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.3778s | 11.7177ms | 85.3413 Ops/s | 33.6096 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 6.5482ms | 2.3089ms | 433.1057 Ops/s | 439.4902 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 6.0958ms | 1.3759ms | 726.7793 Ops/s | 730.0706 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 5.6031ms | 4.3739ms | 228.6271 Ops/s | 221.9735 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 8.8599ms | 2.4341ms | 410.8216 Ops/s | 405.3309 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.2272ms | 1.5648ms | 639.0757 Ops/s | 691.1955 Ops/s | |
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000-100-True] | 44.8301ms | 41.0782ms | 24.3438 Ops/s | 23.7511 Ops/s | |
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000-100-False] | 10.9638ms | 8.8093ms | 113.5167 Ops/s | 114.2709 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_simple | 0.7549s | 0.7421s | 1.3475 Ops/s | 1.3090 Ops/s | |
test_transformed | 0.9779s | 0.9730s | 1.0278 Ops/s | 1.0089 Ops/s | |
test_serial | 2.2193s | 2.1398s | 0.4673 Ops/s | 0.4631 Ops/s | |
test_parallel | 2.0753s | 1.9654s | 0.5088 Ops/s | 0.5071 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.4366ms | 35.7934μs | 27.9381 KOps/s | 27.8773 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 0.3987ms | 21.0290μs | 47.5533 KOps/s | 48.0577 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 53.3310μs | 20.0049μs | 49.9878 KOps/s | 49.6169 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 0.3992ms | 11.7040μs | 85.4406 KOps/s | 84.1989 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 0.4292ms | 38.2461μs | 26.1465 KOps/s | 25.6456 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 57.7710μs | 22.9610μs | 43.5522 KOps/s | 43.7155 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 0.4123ms | 22.0107μs | 45.4325 KOps/s | 44.7550 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 0.4037ms | 14.1487μs | 70.6778 KOps/s | 72.5221 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 0.2339ms | 40.6339μs | 24.6100 KOps/s | 24.9066 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 0.4017ms | 24.8785μs | 40.1954 KOps/s | 40.2623 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 0.4114ms | 21.5102μs | 46.4895 KOps/s | 44.7962 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 55.4910μs | 13.8428μs | 72.2399 KOps/s | 72.6698 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 85.2120μs | 41.5077μs | 24.0919 KOps/s | 23.7594 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 0.4129ms | 26.8486μs | 37.2458 KOps/s | 37.3004 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 59.1210μs | 23.7729μs | 42.0647 KOps/s | 41.8913 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 48.9510μs | 15.7927μs | 63.3203 KOps/s | 64.2527 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 0.4203ms | 40.5938μs | 24.6343 KOps/s | 24.9834 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 0.4043ms | 24.9666μs | 40.0536 KOps/s | 40.3680 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 64.9110μs | 25.7107μs | 38.8943 KOps/s | 38.8957 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 0.3974ms | 15.6366μs | 63.9527 KOps/s | 64.9139 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 0.4271ms | 42.4898μs | 23.5351 KOps/s | 23.9179 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 61.2810μs | 26.9903μs | 37.0504 KOps/s | 37.6156 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 3.2346ms | 28.2117μs | 35.4463 KOps/s | 35.9112 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 0.4083ms | 17.7042μs | 56.4838 KOps/s | 57.5714 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 0.1036ms | 44.3445μs | 22.5507 KOps/s | 22.3546 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 0.4187ms | 28.8937μs | 34.6096 KOps/s | 34.2806 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 0.1402ms | 27.8646μs | 35.8879 KOps/s | 36.1099 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 0.2149ms | 17.7221μs | 56.4266 KOps/s | 57.4767 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 0.2511ms | 45.4055μs | 22.0237 KOps/s | 21.5141 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 78.9510μs | 31.0853μs | 32.1696 KOps/s | 32.3119 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 89.1320μs | 29.5102μs | 33.8866 KOps/s | 34.6843 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 46.5010μs | 19.4022μs | 51.5404 KOps/s | 52.5022 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 24.7211ms | 23.8241ms | 41.9743 Ops/s | 41.7366 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 99.4806ms | 2.8720ms | 348.1850 Ops/s | 325.1004 Ops/s | |
test_values[td0_return_estimate-False-False] | 86.6810μs | 66.2519μs | 15.0939 KOps/s | 15.3618 KOps/s | |
test_values[td1_return_estimate-False-False] | 56.1260ms | 53.5722ms | 18.6664 Ops/s | 18.7355 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 1.3408ms | 1.0680ms | 936.3264 Ops/s | 943.6854 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 89.9217ms | 85.8949ms | 11.6421 Ops/s | 11.7698 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 1.3073ms | 1.0649ms | 939.0278 Ops/s | 942.0925 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.1835ms | 23.9205ms | 41.8052 Ops/s | 42.2436 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.2391ms | 0.7455ms | 1.3414 KOps/s | 1.3697 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7609ms | 0.6487ms | 1.5414 KOps/s | 1.5498 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6251ms | 1.4656ms | 682.3149 Ops/s | 684.4373 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.8389ms | 0.6643ms | 1.5052 KOps/s | 1.5127 KOps/s | |
test_dqn_speed[False-None] | 6.6619ms | 1.2723ms | 785.9775 Ops/s | 783.1977 Ops/s | |
test_dqn_speed[False-backward] | 1.9348ms | 1.7653ms | 566.4849 Ops/s | 563.5175 Ops/s | |
test_dqn_speed[True-None] | 1.1523ms | 0.5549ms | 1.8022 KOps/s | 1.8256 KOps/s | |
test_dqn_speed[True-backward] | 1.1694ms | 1.0137ms | 986.4399 Ops/s | 969.0851 Ops/s | |
test_dqn_speed[reduce-overhead-None] | 0.9886ms | 0.5594ms | 1.7875 KOps/s | 1.8034 KOps/s | |
test_dqn_speed[reduce-overhead-backward] | 1.0911ms | 1.0253ms | 975.3229 Ops/s | 989.6289 Ops/s | |
test_ddpg_speed[False-None] | 3.2082ms | 2.6324ms | 379.8794 Ops/s | 374.4089 Ops/s | |
test_ddpg_speed[False-backward] | 4.0873ms | 3.8509ms | 259.6800 Ops/s | 254.8092 Ops/s | |
test_ddpg_speed[True-None] | 1.6535ms | 1.2318ms | 811.8292 Ops/s | 790.0114 Ops/s | |
test_ddpg_speed[True-backward] | 2.3403ms | 2.1891ms | 456.8074 Ops/s | 451.6116 Ops/s | |
test_ddpg_speed[reduce-overhead-None] | 1.3798ms | 1.2292ms | 813.5345 Ops/s | 806.4841 Ops/s | |
test_ddpg_speed[reduce-overhead-backward] | 2.6199ms | 2.2048ms | 453.5651 Ops/s | 451.0633 Ops/s | |
test_sac_speed[False-None] | 7.9151ms | 7.2963ms | 137.0560 Ops/s | 136.1522 Ops/s | |
test_sac_speed[False-backward] | 11.3746ms | 10.4526ms | 95.6699 Ops/s | 95.5804 Ops/s | |
test_sac_speed[True-None] | 2.4029ms | 2.0048ms | 498.8106 Ops/s | 494.1038 Ops/s | |
test_sac_speed[True-backward] | 4.4687ms | 3.9727ms | 251.7178 Ops/s | 255.9624 Ops/s | |
test_sac_speed[reduce-overhead-None] | 2.2157ms | 1.9938ms | 501.5540 Ops/s | 497.0270 Ops/s | |
test_sac_speed[reduce-overhead-backward] | 4.1697ms | 3.8939ms | 256.8150 Ops/s | 251.8033 Ops/s | |
test_redq_speed[False-None] | 15.7745ms | 11.3357ms | 88.2167 Ops/s | 101.2026 Ops/s | |
test_redq_speed[False-backward] | 17.7510ms | 16.8995ms | 59.1734 Ops/s | 58.3064 Ops/s | |
test_redq_speed[True-None] | 4.3240ms | 3.5323ms | 283.1007 Ops/s | 259.4014 Ops/s | |
test_redq_speed[True-backward] | 8.7830ms | 8.4190ms | 118.7790 Ops/s | 114.6938 Ops/s | |
test_redq_speed[reduce-overhead-None] | 3.7722ms | 3.5439ms | 282.1782 Ops/s | 277.1298 Ops/s | |
test_redq_speed[reduce-overhead-backward] | 8.7618ms | 8.2865ms | 120.6789 Ops/s | 116.7336 Ops/s | |
test_redq_deprec_speed[False-None] | 11.9881ms | 10.1353ms | 98.6652 Ops/s | 94.4596 Ops/s | |
test_redq_deprec_speed[False-backward] | 15.3363ms | 14.8820ms | 67.1954 Ops/s | 64.9871 Ops/s | |
test_redq_deprec_speed[True-None] | 3.4628ms | 3.1844ms | 314.0340 Ops/s | 302.7607 Ops/s | |
test_redq_deprec_speed[True-backward] | 6.9856ms | 6.7521ms | 148.1027 Ops/s | 135.1505 Ops/s | |
test_redq_deprec_speed[reduce-overhead-None] | 3.4951ms | 3.1746ms | 315.0035 Ops/s | 305.2959 Ops/s | |
test_redq_deprec_speed[reduce-overhead-backward] | 6.9733ms | 6.7734ms | 147.6353 Ops/s | 138.4583 Ops/s | |
test_td3_speed[False-None] | 7.4619ms | 7.1717ms | 139.4369 Ops/s | 135.4861 Ops/s | |
test_td3_speed[False-backward] | 10.2317ms | 9.9252ms | 100.7538 Ops/s | 97.2268 Ops/s | |
test_td3_speed[True-None] | 1.9374ms | 1.8844ms | 530.6809 Ops/s | 526.1939 Ops/s | |
test_td3_speed[True-backward] | 3.8412ms | 3.6573ms | 273.4228 Ops/s | 247.5785 Ops/s | |
test_td3_speed[reduce-overhead-None] | 2.0762ms | 1.8938ms | 528.0332 Ops/s | 522.8065 Ops/s | |
test_td3_speed[reduce-overhead-backward] | 3.8996ms | 3.6943ms | 270.6866 Ops/s | 273.8612 Ops/s | |
test_cql_speed[False-None] | 27.4602ms | 24.2601ms | 41.2199 Ops/s | 25.9181 Ops/s | |
test_cql_speed[False-backward] | 36.3645ms | 32.9719ms | 30.3288 Ops/s | 30.0483 Ops/s | |
test_cql_speed[True-None] | 10.8239ms | 10.4277ms | 95.8985 Ops/s | 94.3903 Ops/s | |
test_cql_speed[True-backward] | 16.1202ms | 15.7892ms | 63.3345 Ops/s | 61.5407 Ops/s | |
test_cql_speed[reduce-overhead-None] | 11.1737ms | 10.4670ms | 95.5381 Ops/s | 94.4909 Ops/s | |
test_cql_speed[reduce-overhead-backward] | 16.6379ms | 15.8151ms | 63.2308 Ops/s | 61.2778 Ops/s | |
test_a2c_speed[False-None] | 5.6343ms | 5.1370ms | 194.6646 Ops/s | 193.5976 Ops/s | |
test_a2c_speed[False-backward] | 11.7544ms | 11.4737ms | 87.1556 Ops/s | 86.2044 Ops/s | |
test_a2c_speed[True-None] | 3.1743ms | 2.9804ms | 335.5303 Ops/s | 328.9484 Ops/s | |
test_a2c_speed[True-backward] | 8.4058ms | 8.2348ms | 121.4363 Ops/s | 121.7658 Ops/s | |
test_a2c_speed[reduce-overhead-None] | 3.3091ms | 2.9764ms | 335.9818 Ops/s | 331.4561 Ops/s | |
test_a2c_speed[reduce-overhead-backward] | 8.5138ms | 8.2325ms | 121.4700 Ops/s | 121.2120 Ops/s | |
test_ppo_speed[False-None] | 5.8472ms | 5.4720ms | 182.7489 Ops/s | 182.0759 Ops/s | |
test_ppo_speed[False-backward] | 12.3892ms | 12.0010ms | 83.3267 Ops/s | 84.3200 Ops/s | |
test_ppo_speed[True-None] | 3.5995ms | 3.4148ms | 292.8438 Ops/s | 288.2561 Ops/s | |
test_ppo_speed[True-backward] | 8.2141ms | 8.0098ms | 124.8474 Ops/s | 124.2230 Ops/s | |
test_ppo_speed[reduce-overhead-None] | 3.6195ms | 3.3758ms | 296.2273 Ops/s | 295.6746 Ops/s | |
test_ppo_speed[reduce-overhead-backward] | 8.4291ms | 8.0522ms | 124.1890 Ops/s | 122.2844 Ops/s | |
test_reinforce_speed[False-None] | 5.5317ms | 4.4231ms | 226.0862 Ops/s | 230.5637 Ops/s | |
test_reinforce_speed[False-backward] | 7.3874ms | 7.0998ms | 140.8489 Ops/s | 141.0993 Ops/s | |
test_reinforce_speed[True-None] | 2.5971ms | 2.1498ms | 465.1524 Ops/s | 440.7081 Ops/s | |
test_reinforce_speed[True-backward] | 7.3813ms | 6.9433ms | 144.0231 Ops/s | 135.3048 Ops/s | |
test_reinforce_speed[reduce-overhead-None] | 2.3572ms | 2.1702ms | 460.7948 Ops/s | 458.4889 Ops/s | |
test_reinforce_speed[reduce-overhead-backward] | 7.2249ms | 6.9304ms | 144.2917 Ops/s | 143.8117 Ops/s | |
test_iql_speed[False-None] | 24.5437ms | 19.8128ms | 50.4725 Ops/s | 49.9240 Ops/s | |
test_iql_speed[False-backward] | 35.5296ms | 29.9841ms | 33.3510 Ops/s | 33.1254 Ops/s | |
test_iql_speed[True-None] | 7.0022ms | 6.6845ms | 149.6003 Ops/s | 144.2925 Ops/s | |
test_iql_speed[True-backward] | 15.7041ms | 15.1994ms | 65.7919 Ops/s | 65.2950 Ops/s | |
test_iql_speed[reduce-overhead-None] | 7.2111ms | 6.6952ms | 149.3601 Ops/s | 148.9045 Ops/s | |
test_iql_speed[reduce-overhead-backward] | 15.8388ms | 15.2698ms | 65.4886 Ops/s | 64.6712 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.4853ms | 6.2675ms | 159.5542 Ops/s | 156.7024 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.5660ms | 0.2807ms | 3.5621 KOps/s | 3.2547 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6553ms | 0.2413ms | 4.1436 KOps/s | 3.3298 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.4250ms | 6.0879ms | 164.2615 Ops/s | 164.7564 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 2.2774ms | 0.2496ms | 4.0063 KOps/s | 3.5862 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.4286ms | 0.2289ms | 4.3690 KOps/s | 3.1510 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.5854ms | 1.1968ms | 835.5851 Ops/s | 726.4520 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5416ms | 1.1529ms | 867.3904 Ops/s | 736.1949 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.4759ms | 6.2556ms | 159.8568 Ops/s | 160.2113 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.7776ms | 0.4048ms | 2.4703 KOps/s | 2.1114 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7600ms | 0.3803ms | 2.6297 KOps/s | 2.3739 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.3727ms | 6.1570ms | 162.4170 Ops/s | 163.4854 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.5397ms | 0.2579ms | 3.8772 KOps/s | 3.7995 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.4768ms | 0.2356ms | 4.2439 KOps/s | 4.2049 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.4624ms | 6.0170ms | 166.1944 Ops/s | 165.4015 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6621ms | 0.2556ms | 3.9117 KOps/s | 2.5550 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5783ms | 0.2371ms | 4.2170 KOps/s | 2.6674 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 10.1847ms | 6.6085ms | 151.3198 Ops/s | 160.0605 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.4039s | 0.9101ms | 1.0988 KOps/s | 2.1864 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.5950ms | 0.3738ms | 2.6754 KOps/s | 2.1483 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 7.0257ms | 5.2964ms | 188.8084 Ops/s | 193.0936 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 9.4837ms | 2.0300ms | 492.6207 Ops/s | 467.8042 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 7.1891ms | 1.2183ms | 820.8451 Ops/s | 777.4048 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.3784s | 12.7259ms | 78.5797 Ops/s | 33.2932 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 9.9168ms | 2.0129ms | 496.7884 Ops/s | 488.6275 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 3.3369ms | 1.1093ms | 901.4845 Ops/s | 834.4069 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 10.0279ms | 5.5415ms | 180.4560 Ops/s | 177.8473 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 6.8790ms | 2.1271ms | 470.1174 Ops/s | 454.9228 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 9.1514ms | 1.3891ms | 719.9032 Ops/s | 691.4633 Ops/s | |
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000-100-True] | 47.4459ms | 46.1232ms | 21.6811 Ops/s | 21.0139 Ops/s | |
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000-100-False] | 10.0377ms | 9.4809ms | 105.4757 Ops/s | 103.5725 Ops/s |
vmoens
added a commit
that referenced
this pull request
Nov 4, 2024
ghstack-source-id: c63a63b16696a323715d9e98939b990b90a43177 Pull Request resolved: #2532
vmoens
added a commit
that referenced
this pull request
Nov 4, 2024
ghstack-source-id: 62a3091e5c9072f26266143319f30de1729c0d4e Pull Request resolved: #2532
vmoens
added a commit
that referenced
this pull request
Nov 4, 2024
ghstack-source-id: 62a3091e5c9072f26266143319f30de1729c0d4e Pull Request resolved: #2532
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.
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):