-
Notifications
You must be signed in to change notification settings - Fork 303
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] inline hold_out_net
#2498
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/2498
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 6 Unrelated FailuresAs of commit ad3c280 with merge base d894358 (): NEW FAILURE - The following job has 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
Oct 17, 2024
ghstack-source-id: c315202c8af55f0852195fe488ae855966386c4c Pull Request resolved: #2498
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
Oct 17, 2024
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_simple | 0.4161s | 0.4141s | 2.4147 Ops/s | 2.3688 Ops/s | |
test_transformed | 0.6879s | 0.6011s | 1.6637 Ops/s | 1.6592 Ops/s | |
test_serial | 1.4212s | 1.3399s | 0.7463 Ops/s | 0.7247 Ops/s | |
test_parallel | 1.3335s | 1.3172s | 0.7592 Ops/s | 0.7421 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1370ms | 29.5487μs | 33.8425 KOps/s | 34.0146 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 38.4420μs | 17.5742μs | 56.9017 KOps/s | 57.9243 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 85.4490μs | 16.3319μs | 61.2299 KOps/s | 60.6336 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 38.4320μs | 9.6987μs | 103.1063 KOps/s | 103.7778 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 0.1086ms | 31.9378μs | 31.3108 KOps/s | 31.4405 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 54.5510μs | 19.8639μs | 50.3427 KOps/s | 51.3155 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 80.5100μs | 18.6509μs | 53.6168 KOps/s | 53.6234 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 40.3040μs | 11.8236μs | 84.5768 KOps/s | 84.3091 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 89.6770μs | 33.8693μs | 29.5253 KOps/s | 29.7798 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 0.6267ms | 22.0321μs | 45.3883 KOps/s | 45.9209 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 73.2960μs | 18.6115μs | 53.7301 KOps/s | 54.1938 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 55.5130μs | 11.6160μs | 86.0878 KOps/s | 84.8315 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 0.1083ms | 35.6226μs | 28.0720 KOps/s | 28.0541 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 76.4320μs | 24.0214μs | 41.6296 KOps/s | 42.2307 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 64.9710μs | 20.4367μs | 48.9315 KOps/s | 48.8602 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 76.2520μs | 13.8051μs | 72.4368 KOps/s | 72.6486 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 70.9020μs | 33.8871μs | 29.5098 KOps/s | 29.6427 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 77.3440μs | 21.8068μs | 45.8573 KOps/s | 45.7035 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 55.1030μs | 21.5545μs | 46.3939 KOps/s | 45.1379 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 66.9940μs | 13.5153μs | 73.9902 KOps/s | 73.3821 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 74.1170μs | 35.8371μs | 27.9041 KOps/s | 28.0263 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 77.1630μs | 23.9103μs | 41.8230 KOps/s | 42.2790 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 3.1575ms | 23.9094μs | 41.8245 KOps/s | 41.8241 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 83.3350μs | 15.6830μs | 63.7631 KOps/s | 64.1145 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 85.1080μs | 38.0172μs | 26.3039 KOps/s | 26.6491 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 59.2100μs | 26.1302μs | 38.2699 KOps/s | 38.5179 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 73.5060μs | 24.3494μs | 41.0688 KOps/s | 42.5057 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 45.0940μs | 15.5531μs | 64.2960 KOps/s | 63.9905 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 0.6725ms | 39.7733μs | 25.1425 KOps/s | 25.3002 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 99.4510μs | 27.5801μs | 36.2580 KOps/s | 35.9196 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 64.9200μs | 24.8980μs | 40.1639 KOps/s | 38.7231 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 53.9610μs | 17.2201μs | 58.0716 KOps/s | 56.8088 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 10.4963ms | 9.4298ms | 106.0472 Ops/s | 104.0753 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 36.4737ms | 33.6656ms | 29.7039 Ops/s | 29.6545 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.3599ms | 0.1997ms | 5.0087 KOps/s | 5.8200 KOps/s | |
test_values[td1_return_estimate-False-False] | 31.8813ms | 24.3156ms | 41.1259 Ops/s | 41.9684 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 36.3439ms | 33.7852ms | 29.5987 Ops/s | 29.4504 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 39.2474ms | 34.8835ms | 28.6668 Ops/s | 29.0340 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 35.8607ms | 33.7041ms | 29.6700 Ops/s | 29.6158 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 12.3438ms | 8.1936ms | 122.0469 Ops/s | 119.9855 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.4244ms | 2.0342ms | 491.5912 Ops/s | 501.5074 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.5586ms | 0.3537ms | 2.8273 KOps/s | 2.7734 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 47.0890ms | 46.3088ms | 21.5942 Ops/s | 21.2179 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.3017ms | 3.0498ms | 327.8947 Ops/s | 323.9125 Ops/s | |
test_dqn_speed[False-None] | 1.7263ms | 1.3363ms | 748.3322 Ops/s | 714.7884 Ops/s | |
test_dqn_speed[False-backward] | 2.1294ms | 1.8451ms | 541.9831 Ops/s | 537.6510 Ops/s | |
test_dqn_speed[True-None] | 0.6595ms | 0.4589ms | 2.1790 KOps/s | 2.1305 KOps/s | |
test_dqn_speed[True-backward] | 0.9706ms | 0.8749ms | 1.1429 KOps/s | 1.1211 KOps/s | |
test_dqn_speed[reduce-overhead-None] | 0.7070ms | 0.4654ms | 2.1487 KOps/s | 2.1328 KOps/s | |
test_dqn_speed[reduce-overhead-backward] | 0.9420ms | 0.8857ms | 1.1290 KOps/s | 1.1450 KOps/s | |
test_ddpg_speed[False-None] | 4.1297ms | 2.7864ms | 358.8832 Ops/s | 348.4982 Ops/s | |
test_ddpg_speed[False-backward] | 4.2232ms | 3.9015ms | 256.3123 Ops/s | 251.2504 Ops/s | |
test_ddpg_speed[True-None] | 1.3346ms | 0.9983ms | 1.0018 KOps/s | 992.5087 Ops/s | |
test_ddpg_speed[True-backward] | 2.0611ms | 1.8820ms | 531.3558 Ops/s | 502.7271 Ops/s | |
test_ddpg_speed[reduce-overhead-None] | 1.2782ms | 1.0019ms | 998.1020 Ops/s | 962.8307 Ops/s | |
test_ddpg_speed[reduce-overhead-backward] | 2.1071ms | 1.9014ms | 525.9327 Ops/s | 521.5434 Ops/s | |
test_sac_speed[False-None] | 8.9232ms | 7.9731ms | 125.4210 Ops/s | 121.4824 Ops/s | |
test_sac_speed[False-backward] | 13.4518ms | 10.7418ms | 93.0941 Ops/s | 89.8491 Ops/s | |
test_sac_speed[True-None] | 2.9885ms | 1.8533ms | 539.5845 Ops/s | 535.8181 Ops/s | |
test_sac_speed[True-backward] | 3.8703ms | 3.5428ms | 282.2592 Ops/s | 283.9220 Ops/s | |
test_sac_speed[reduce-overhead-None] | 2.4873ms | 1.8807ms | 531.7136 Ops/s | 522.2129 Ops/s | |
test_sac_speed[reduce-overhead-backward] | 3.7430ms | 3.5690ms | 280.1936 Ops/s | 277.4346 Ops/s | |
test_redq_speed[False-None] | 14.8044ms | 13.1022ms | 76.3231 Ops/s | 77.3390 Ops/s | |
test_redq_speed[False-backward] | 23.8131ms | 22.2608ms | 44.9221 Ops/s | 43.9406 Ops/s | |
test_redq_speed[True-None] | 5.7824ms | 5.0457ms | 198.1870 Ops/s | 201.5561 Ops/s | |
test_redq_speed[True-backward] | 13.5684ms | 12.8136ms | 78.0422 Ops/s | 79.4627 Ops/s | |
test_redq_speed[reduce-overhead-None] | 6.5654ms | 4.9914ms | 200.3438 Ops/s | 207.2214 Ops/s | |
test_redq_speed[reduce-overhead-backward] | 14.5408ms | 12.3878ms | 80.7244 Ops/s | 80.4477 Ops/s | |
test_redq_deprec_speed[False-None] | 17.5950ms | 13.1922ms | 75.8024 Ops/s | 76.2168 Ops/s | |
test_redq_deprec_speed[False-backward] | 21.3750ms | 18.8800ms | 52.9662 Ops/s | 52.6615 Ops/s | |
test_redq_deprec_speed[True-None] | 4.7378ms | 3.7571ms | 266.1604 Ops/s | 273.5297 Ops/s | |
test_redq_deprec_speed[True-backward] | 9.3176ms | 8.5459ms | 117.0151 Ops/s | 120.3562 Ops/s | |
test_redq_deprec_speed[reduce-overhead-None] | 4.4618ms | 3.6715ms | 272.3690 Ops/s | 275.9369 Ops/s | |
test_redq_deprec_speed[reduce-overhead-backward] | 8.9333ms | 8.4971ms | 117.6866 Ops/s | 107.7252 Ops/s | |
test_td3_speed[False-None] | 8.4824ms | 7.8674ms | 127.1064 Ops/s | 121.8969 Ops/s | |
test_td3_speed[False-backward] | 12.6613ms | 10.7264ms | 93.2276 Ops/s | 93.5737 Ops/s | |
test_td3_speed[True-None] | 1.9907ms | 1.7365ms | 575.8639 Ops/s | 566.3217 Ops/s | |
test_td3_speed[True-backward] | 3.5248ms | 3.3523ms | 298.3038 Ops/s | 295.1879 Ops/s | |
test_td3_speed[reduce-overhead-None] | 1.8415ms | 1.7332ms | 576.9556 Ops/s | 561.3855 Ops/s | |
test_td3_speed[reduce-overhead-backward] | 3.5316ms | 3.3450ms | 298.9520 Ops/s | 286.9940 Ops/s | |
test_cql_speed[False-None] | 37.5092ms | 35.5191ms | 28.1539 Ops/s | 27.4384 Ops/s | |
test_cql_speed[False-backward] | 47.5973ms | 45.6095ms | 21.9252 Ops/s | 21.3887 Ops/s | |
test_cql_speed[True-None] | 17.4069ms | 15.9001ms | 62.8928 Ops/s | 60.9908 Ops/s | |
test_cql_speed[True-backward] | 24.2412ms | 22.8718ms | 43.7220 Ops/s | 42.7284 Ops/s | |
test_cql_speed[reduce-overhead-None] | 17.4027ms | 16.0626ms | 62.2564 Ops/s | 61.2307 Ops/s | |
test_cql_speed[reduce-overhead-backward] | 24.9738ms | 23.1071ms | 43.2767 Ops/s | 44.0611 Ops/s | |
test_a2c_speed[False-None] | 8.2729ms | 7.5052ms | 133.2415 Ops/s | 133.3703 Ops/s | |
test_a2c_speed[False-backward] | 15.8464ms | 15.0714ms | 66.3509 Ops/s | 66.4428 Ops/s | |
test_a2c_speed[True-None] | 3.7447ms | 3.3925ms | 294.7685 Ops/s | 295.0607 Ops/s | |
test_a2c_speed[True-backward] | 10.9028ms | 10.2901ms | 97.1803 Ops/s | 99.7268 Ops/s | |
test_a2c_speed[reduce-overhead-None] | 4.1256ms | 3.4797ms | 287.3824 Ops/s | 291.5859 Ops/s | |
test_a2c_speed[reduce-overhead-backward] | 10.7855ms | 10.4293ms | 95.8838 Ops/s | 99.8256 Ops/s | |
test_ppo_speed[False-None] | 8.5799ms | 7.8533ms | 127.3353 Ops/s | 132.4403 Ops/s | |
test_ppo_speed[False-backward] | 16.7943ms | 15.5018ms | 64.5085 Ops/s | 65.9839 Ops/s | |
test_ppo_speed[True-None] | 4.4923ms | 3.7309ms | 268.0300 Ops/s | 266.6451 Ops/s | |
test_ppo_speed[True-backward] | 10.9982ms | 10.0683ms | 99.3217 Ops/s | 98.4043 Ops/s | |
test_ppo_speed[reduce-overhead-None] | 4.0575ms | 3.7225ms | 268.6340 Ops/s | 264.6415 Ops/s | |
test_ppo_speed[reduce-overhead-backward] | 10.5590ms | 9.9186ms | 100.8205 Ops/s | 101.7803 Ops/s | |
test_reinforce_speed[False-None] | 7.5758ms | 6.5813ms | 151.9446 Ops/s | 151.0512 Ops/s | |
test_reinforce_speed[False-backward] | 11.3398ms | 10.0938ms | 99.0706 Ops/s | 99.9132 Ops/s | |
test_reinforce_speed[True-None] | 3.0759ms | 2.6831ms | 372.7016 Ops/s | 370.7369 Ops/s | |
test_reinforce_speed[True-backward] | 9.8319ms | 9.0416ms | 110.5995 Ops/s | 113.3616 Ops/s | |
test_reinforce_speed[reduce-overhead-None] | 3.0437ms | 2.7547ms | 363.0128 Ops/s | 363.9582 Ops/s | |
test_reinforce_speed[reduce-overhead-backward] | 10.6045ms | 8.9815ms | 111.3400 Ops/s | 111.0116 Ops/s | |
test_iql_speed[False-None] | 33.9599ms | 32.4858ms | 30.7827 Ops/s | 30.7157 Ops/s | |
test_iql_speed[False-backward] | 54.4703ms | 45.8698ms | 21.8008 Ops/s | 21.7428 Ops/s | |
test_iql_speed[True-None] | 12.2447ms | 10.9361ms | 91.4399 Ops/s | 89.9486 Ops/s | |
test_iql_speed[True-backward] | 23.9086ms | 22.7386ms | 43.9781 Ops/s | 43.9321 Ops/s | |
test_iql_speed[reduce-overhead-None] | 11.9969ms | 11.0179ms | 90.7613 Ops/s | 90.6083 Ops/s | |
test_iql_speed[reduce-overhead-backward] | 23.6448ms | 22.4313ms | 44.5805 Ops/s | 43.4647 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 5.2540ms | 4.9038ms | 203.9248 Ops/s | 199.9888 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.2010ms | 0.4855ms | 2.0595 KOps/s | 2.0300 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6992ms | 0.4644ms | 2.1534 KOps/s | 2.1347 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.2972ms | 4.9148ms | 203.4671 Ops/s | 202.1651 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 2.7785ms | 0.4846ms | 2.0638 KOps/s | 2.0678 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7557ms | 0.4677ms | 2.1379 KOps/s | 2.1512 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.8835ms | 1.5874ms | 629.9771 Ops/s | 621.5569 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.8248ms | 1.5312ms | 653.0948 Ops/s | 641.1692 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 7.4644ms | 5.0307ms | 198.7795 Ops/s | 194.9257 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.4480s | 0.9724ms | 1.0284 KOps/s | 1.5808 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8462ms | 0.5936ms | 1.6847 KOps/s | 1.5778 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.3035ms | 4.8461ms | 206.3522 Ops/s | 200.7369 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.7194ms | 0.4902ms | 2.0402 KOps/s | 2.0033 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7053ms | 0.4626ms | 2.1619 KOps/s | 2.1350 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.7734ms | 4.8664ms | 205.4920 Ops/s | 204.3209 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.0518ms | 0.4723ms | 2.1173 KOps/s | 2.0748 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6421ms | 0.4599ms | 2.1744 KOps/s | 2.0451 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.2469ms | 4.9335ms | 202.6966 Ops/s | 198.3768 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.3523ms | 0.6246ms | 1.6010 KOps/s | 1.5997 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 1.0248ms | 0.6344ms | 1.5763 KOps/s | 1.6391 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4278s | 12.7225ms | 78.6007 Ops/s | 247.1572 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 9.1245ms | 2.4352ms | 410.6490 Ops/s | 435.7144 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 2.3364ms | 1.3008ms | 768.7416 Ops/s | 709.7301 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 5.4617ms | 4.1473ms | 241.1190 Ops/s | 33.6762 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 5.7600ms | 2.2614ms | 442.2068 Ops/s | 426.9065 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 6.2173ms | 1.3409ms | 745.7811 Ops/s | 763.4544 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.4016s | 12.3173ms | 81.1869 Ops/s | 220.1523 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 8.9126ms | 2.5157ms | 397.5057 Ops/s | 386.2671 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 6.2144ms | 1.5021ms | 665.7129 Ops/s | 665.7494 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_simple | 0.7212s | 0.7203s | 1.3883 Ops/s | 1.3757 Ops/s | |
test_transformed | 1.0602s | 0.9807s | 1.0197 Ops/s | 1.0291 Ops/s | |
test_serial | 2.2729s | 2.1653s | 0.4618 Ops/s | 0.4702 Ops/s | |
test_parallel | 2.0467s | 1.9845s | 0.5039 Ops/s | 0.5033 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1999ms | 39.5587μs | 25.2789 KOps/s | 25.1786 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 0.2220ms | 23.5744μs | 42.4189 KOps/s | 42.3866 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 92.7220μs | 21.6403μs | 46.2100 KOps/s | 45.9414 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 38.3600μs | 12.7768μs | 78.2666 KOps/s | 78.9355 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 77.6210μs | 42.2909μs | 23.6458 KOps/s | 23.8690 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 62.0210μs | 25.4211μs | 39.3373 KOps/s | 39.0361 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 53.7210μs | 24.4896μs | 40.8337 KOps/s | 40.0924 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 42.8900μs | 15.2357μs | 65.6353 KOps/s | 65.0684 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 95.0910μs | 45.5892μs | 21.9350 KOps/s | 21.9836 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 64.0000μs | 28.6816μs | 34.8655 KOps/s | 34.6480 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 0.1662ms | 24.2718μs | 41.2000 KOps/s | 39.7545 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 0.1065ms | 15.2276μs | 65.6702 KOps/s | 64.5904 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 77.3710μs | 47.4290μs | 21.0841 KOps/s | 20.7006 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 65.3410μs | 30.7722μs | 32.4969 KOps/s | 31.5809 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 61.2010μs | 26.8572μs | 37.2340 KOps/s | 35.9841 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 50.3010μs | 17.6989μs | 56.5007 KOps/s | 55.3479 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 74.6010μs | 44.6447μs | 22.3991 KOps/s | 21.9039 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 0.1142ms | 28.1842μs | 35.4808 KOps/s | 34.4044 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 83.9910μs | 28.6276μs | 34.9313 KOps/s | 33.6374 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 51.2600μs | 17.7731μs | 56.2649 KOps/s | 55.6088 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 85.1010μs | 47.1061μs | 21.2287 KOps/s | 20.4960 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 64.6510μs | 30.6889μs | 32.5851 KOps/s | 31.8698 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 3.5064ms | 31.7124μs | 31.5334 KOps/s | 30.7968 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 51.3010μs | 20.0351μs | 49.9123 KOps/s | 48.8655 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 87.6010μs | 49.8841μs | 20.0465 KOps/s | 19.5710 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 61.2910μs | 33.2004μs | 30.1201 KOps/s | 29.4179 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 69.5210μs | 31.7281μs | 31.5178 KOps/s | 31.3196 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 49.7710μs | 20.0388μs | 49.9031 KOps/s | 48.8232 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 84.4920μs | 51.1188μs | 19.5623 KOps/s | 18.8178 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 64.3910μs | 35.7908μs | 27.9401 KOps/s | 27.2762 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 69.8510μs | 32.9337μs | 30.3641 KOps/s | 29.5464 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 51.4500μs | 22.4012μs | 44.6405 KOps/s | 44.1534 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 23.8170ms | 23.4631ms | 42.6202 Ops/s | 42.4781 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 98.5944ms | 2.8435ms | 351.6846 Ops/s | 348.2456 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.1067ms | 64.0353μs | 15.6164 KOps/s | 15.9594 KOps/s | |
test_values[td1_return_estimate-False-False] | 52.7155ms | 52.3173ms | 19.1141 Ops/s | 19.0575 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 1.2230ms | 1.0517ms | 950.8004 Ops/s | 946.6593 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 84.2217ms | 83.1142ms | 12.0316 Ops/s | 12.0398 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 1.1789ms | 1.0386ms | 962.7890 Ops/s | 951.3495 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 23.5878ms | 23.2097ms | 43.0854 Ops/s | 43.1603 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.0322ms | 0.7161ms | 1.3964 KOps/s | 1.4094 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7678ms | 0.6318ms | 1.5828 KOps/s | 1.5755 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6034ms | 1.4454ms | 691.8601 Ops/s | 691.7729 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.7980ms | 0.6473ms | 1.5448 KOps/s | 1.5416 KOps/s | |
test_dqn_speed[False-None] | 6.8774ms | 1.3214ms | 756.7932 Ops/s | 680.5734 Ops/s | |
test_dqn_speed[False-backward] | 2.0016ms | 1.8488ms | 540.8928 Ops/s | 553.9601 Ops/s | |
test_dqn_speed[True-None] | 1.2056ms | 0.5512ms | 1.8141 KOps/s | 1.7654 KOps/s | |
test_dqn_speed[True-backward] | 1.2558ms | 1.0212ms | 979.2671 Ops/s | 836.0439 Ops/s | |
test_dqn_speed[reduce-overhead-None] | 0.8553ms | 0.5549ms | 1.8020 KOps/s | 1.7648 KOps/s | |
test_dqn_speed[reduce-overhead-backward] | 1.1070ms | 1.0207ms | 979.7350 Ops/s | 977.1599 Ops/s | |
test_ddpg_speed[False-None] | 3.0300ms | 2.6882ms | 371.9909 Ops/s | 358.3694 Ops/s | |
test_ddpg_speed[False-backward] | 4.1605ms | 3.9910ms | 250.5667 Ops/s | 252.4905 Ops/s | |
test_ddpg_speed[True-None] | 1.6068ms | 1.2547ms | 796.9868 Ops/s | 772.1505 Ops/s | |
test_ddpg_speed[True-backward] | 2.3438ms | 2.2240ms | 449.6321 Ops/s | 444.4185 Ops/s | |
test_ddpg_speed[reduce-overhead-None] | 1.6086ms | 1.2513ms | 799.1561 Ops/s | 792.7412 Ops/s | |
test_ddpg_speed[reduce-overhead-backward] | 2.5406ms | 2.2413ms | 446.1639 Ops/s | 450.9101 Ops/s | |
test_sac_speed[False-None] | 7.9915ms | 7.5684ms | 132.1282 Ops/s | 129.5323 Ops/s | |
test_sac_speed[False-backward] | 11.3020ms | 10.7462ms | 93.0560 Ops/s | 92.1105 Ops/s | |
test_sac_speed[True-None] | 2.4139ms | 2.0560ms | 486.3879 Ops/s | 487.9698 Ops/s | |
test_sac_speed[True-backward] | 4.3392ms | 4.0353ms | 247.8127 Ops/s | 212.1828 Ops/s | |
test_sac_speed[reduce-overhead-None] | 2.3147ms | 2.0679ms | 483.5933 Ops/s | 488.4371 Ops/s | |
test_sac_speed[reduce-overhead-backward] | 4.1768ms | 3.9991ms | 250.0546 Ops/s | 250.0533 Ops/s | |
test_redq_speed[False-None] | 15.5342ms | 10.5776ms | 94.5397 Ops/s | 87.4636 Ops/s | |
test_redq_speed[False-backward] | 19.3372ms | 17.8183ms | 56.1220 Ops/s | 54.4975 Ops/s | |
test_redq_speed[True-None] | 4.2672ms | 3.7750ms | 264.9042 Ops/s | 220.2820 Ops/s | |
test_redq_speed[True-backward] | 9.2259ms | 8.8074ms | 113.5413 Ops/s | 106.6975 Ops/s | |
test_redq_speed[reduce-overhead-None] | 3.8961ms | 3.6384ms | 274.8449 Ops/s | 275.6581 Ops/s | |
test_redq_speed[reduce-overhead-backward] | 9.2712ms | 8.8346ms | 113.1909 Ops/s | 113.2151 Ops/s | |
test_redq_deprec_speed[False-None] | 11.2818ms | 10.7566ms | 92.9665 Ops/s | 94.1951 Ops/s | |
test_redq_deprec_speed[False-backward] | 16.0399ms | 15.5789ms | 64.1894 Ops/s | 65.0372 Ops/s | |
test_redq_deprec_speed[True-None] | 3.6273ms | 3.3013ms | 302.9097 Ops/s | 286.5308 Ops/s | |
test_redq_deprec_speed[True-backward] | 7.6590ms | 7.2204ms | 138.4959 Ops/s | 136.2540 Ops/s | |
test_redq_deprec_speed[reduce-overhead-None] | 3.7969ms | 3.3335ms | 299.9852 Ops/s | 299.6082 Ops/s | |
test_redq_deprec_speed[reduce-overhead-backward] | 7.5455ms | 7.2533ms | 137.8686 Ops/s | 136.1580 Ops/s | |
test_td3_speed[False-None] | 7.7213ms | 7.5347ms | 132.7188 Ops/s | 131.5927 Ops/s | |
test_td3_speed[False-backward] | 10.7714ms | 10.3559ms | 96.5630 Ops/s | 95.9742 Ops/s | |
test_td3_speed[True-None] | 1.9927ms | 1.9409ms | 515.2159 Ops/s | 512.0085 Ops/s | |
test_td3_speed[True-backward] | 3.9858ms | 3.8360ms | 260.6863 Ops/s | 237.9496 Ops/s | |
test_td3_speed[reduce-overhead-None] | 2.0893ms | 1.9600ms | 510.2144 Ops/s | 517.7715 Ops/s | |
test_td3_speed[reduce-overhead-backward] | 3.9189ms | 3.7998ms | 263.1717 Ops/s | 262.4117 Ops/s | |
test_cql_speed[False-None] | 27.7563ms | 25.5378ms | 39.1577 Ops/s | 39.9979 Ops/s | |
test_cql_speed[False-backward] | 39.7691ms | 35.0446ms | 28.5351 Ops/s | 29.4977 Ops/s | |
test_cql_speed[True-None] | 11.7754ms | 11.2094ms | 89.2106 Ops/s | 92.5706 Ops/s | |
test_cql_speed[True-backward] | 17.7229ms | 17.0581ms | 58.6231 Ops/s | 60.2636 Ops/s | |
test_cql_speed[reduce-overhead-None] | 14.9229ms | 11.4425ms | 87.3934 Ops/s | 90.0990 Ops/s | |
test_cql_speed[reduce-overhead-backward] | 17.5704ms | 17.1275ms | 58.3858 Ops/s | 60.1274 Ops/s | |
test_a2c_speed[False-None] | 7.4464ms | 5.3299ms | 187.6192 Ops/s | 186.1896 Ops/s | |
test_a2c_speed[False-backward] | 12.4034ms | 11.8838ms | 84.1481 Ops/s | 85.0195 Ops/s | |
test_a2c_speed[True-None] | 3.5268ms | 3.1436ms | 318.1023 Ops/s | 319.9285 Ops/s | |
test_a2c_speed[True-backward] | 9.0550ms | 8.6005ms | 116.2719 Ops/s | 95.7701 Ops/s | |
test_a2c_speed[reduce-overhead-None] | 3.3578ms | 3.1176ms | 320.7568 Ops/s | 314.8480 Ops/s | |
test_a2c_speed[reduce-overhead-backward] | 8.8063ms | 8.5311ms | 117.2176 Ops/s | 116.4370 Ops/s | |
test_ppo_speed[False-None] | 6.0061ms | 5.7381ms | 174.2743 Ops/s | 172.0933 Ops/s | |
test_ppo_speed[False-backward] | 12.9044ms | 12.5558ms | 79.6445 Ops/s | 80.1900 Ops/s | |
test_ppo_speed[True-None] | 3.6822ms | 3.4939ms | 286.2105 Ops/s | 284.5195 Ops/s | |
test_ppo_speed[True-backward] | 8.6300ms | 8.3066ms | 120.3857 Ops/s | 119.9866 Ops/s | |
test_ppo_speed[reduce-overhead-None] | 3.7587ms | 3.4977ms | 285.8984 Ops/s | 285.4431 Ops/s | |
test_ppo_speed[reduce-overhead-backward] | 8.6080ms | 8.2723ms | 120.8851 Ops/s | 120.3947 Ops/s | |
test_reinforce_speed[False-None] | 6.5676ms | 4.5024ms | 222.1022 Ops/s | 219.3120 Ops/s | |
test_reinforce_speed[False-backward] | 7.6041ms | 7.3020ms | 136.9481 Ops/s | 136.6380 Ops/s | |
test_reinforce_speed[True-None] | 2.6017ms | 2.2873ms | 437.1987 Ops/s | 429.1368 Ops/s | |
test_reinforce_speed[True-backward] | 7.4879ms | 7.1822ms | 139.2336 Ops/s | 136.6537 Ops/s | |
test_reinforce_speed[reduce-overhead-None] | 2.6927ms | 2.2759ms | 439.3892 Ops/s | 440.4454 Ops/s | |
test_reinforce_speed[reduce-overhead-backward] | 7.6413ms | 7.1966ms | 138.9548 Ops/s | 138.0501 Ops/s | |
test_iql_speed[False-None] | 19.6947ms | 19.0355ms | 52.5335 Ops/s | 50.4365 Ops/s | |
test_iql_speed[False-backward] | 35.4559ms | 30.1845ms | 33.1295 Ops/s | 33.0469 Ops/s | |
test_iql_speed[True-None] | 7.5131ms | 6.9142ms | 144.6298 Ops/s | 142.5205 Ops/s | |
test_iql_speed[True-backward] | 16.7706ms | 15.8917ms | 62.9261 Ops/s | 61.9420 Ops/s | |
test_iql_speed[reduce-overhead-None] | 8.1160ms | 7.0709ms | 141.4247 Ops/s | 145.6343 Ops/s | |
test_iql_speed[reduce-overhead-backward] | 16.6603ms | 15.8924ms | 62.9231 Ops/s | 63.1140 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.3858ms | 6.1816ms | 161.7704 Ops/s | 162.4975 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.2590ms | 0.2682ms | 3.7282 KOps/s | 4.2027 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.5372ms | 0.2804ms | 3.5669 KOps/s | 4.6292 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.5001ms | 6.1283ms | 163.1783 Ops/s | 165.3807 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 2.0109ms | 0.2709ms | 3.6915 KOps/s | 2.8604 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5453ms | 0.2804ms | 3.5669 KOps/s | 3.0040 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.4283ms | 1.1963ms | 835.8935 Ops/s | 710.2511 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.6339ms | 1.1450ms | 873.3288 Ops/s | 726.1611 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.5113ms | 6.2614ms | 159.7088 Ops/s | 160.8625 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.2369ms | 0.3815ms | 2.6212 KOps/s | 2.3271 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.6181ms | 0.3583ms | 2.7907 KOps/s | 2.4636 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.3809ms | 6.0967ms | 164.0229 Ops/s | 164.6405 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.4868s | 0.7726ms | 1.2943 KOps/s | 3.6157 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.4546ms | 0.2253ms | 4.4394 KOps/s | 4.1622 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.4486ms | 6.1180ms | 163.4520 Ops/s | 161.8525 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.7968ms | 0.2376ms | 4.2093 KOps/s | 4.1218 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5849ms | 0.2167ms | 4.6139 KOps/s | 4.4231 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.4712ms | 6.3099ms | 158.4806 Ops/s | 156.7927 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9636ms | 0.3839ms | 2.6048 KOps/s | 2.5709 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.5199ms | 0.3628ms | 2.7562 KOps/s | 2.7413 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 6.7491ms | 5.2126ms | 191.8418 Ops/s | 189.6821 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 6.5990ms | 2.1754ms | 459.6880 Ops/s | 455.0754 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 8.2906ms | 1.2617ms | 792.5694 Ops/s | 813.7864 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.4142s | 13.3879ms | 74.6941 Ops/s | 189.3031 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 9.0670ms | 2.0215ms | 494.6732 Ops/s | 448.9575 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 7.5261ms | 1.2204ms | 819.4036 Ops/s | 835.3060 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.4031s | 13.3143ms | 75.1074 Ops/s | 186.1435 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 9.5473ms | 2.1833ms | 458.0310 Ops/s | 414.3013 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 6.9006ms | 1.3701ms | 729.8906 Ops/s | 741.2467 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.
enhancement
New feature or request
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):
hold_out_net
#2498