-
Notifications
You must be signed in to change notification settings - Fork 326
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
[Examples] Move examples to sota-implementations #2016
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2016
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 2 Unrelated FailuresAs of commit 9ffeea2 with merge base 29d9a5b (): 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. |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 55.9195ms | 55.5346ms | 18.0068 Ops/s | 17.0450 Ops/s | |
test_sync | 36.9993ms | 31.7950ms | 31.4515 Ops/s | 30.3325 Ops/s | |
test_async | 53.0661ms | 27.2438ms | 36.7057 Ops/s | 34.4595 Ops/s | |
test_simple | 0.4166s | 0.3543s | 2.8228 Ops/s | 2.8646 Ops/s | |
test_transformed | 0.5308s | 0.4802s | 2.0826 Ops/s | 2.1355 Ops/s | |
test_serial | 1.2523s | 1.2089s | 0.8272 Ops/s | 0.8219 Ops/s | |
test_parallel | 1.0991s | 1.0558s | 0.9472 Ops/s | 0.9543 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 53.1290μs | 21.5193μs | 46.4700 KOps/s | 46.4903 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 37.0190μs | 13.2449μs | 75.5005 KOps/s | 77.4824 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 44.5530μs | 12.5662μs | 79.5786 KOps/s | 79.6848 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 33.5430μs | 7.6985μs | 129.8952 KOps/s | 132.4992 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 54.4320μs | 22.8070μs | 43.8462 KOps/s | 43.9209 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 38.2520μs | 14.5070μs | 68.9320 KOps/s | 70.5373 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 34.4750μs | 13.8943μs | 71.9720 KOps/s | 73.1702 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 39.1130μs | 9.0529μs | 110.4622 KOps/s | 114.0958 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 0.1527ms | 24.7348μs | 40.4288 KOps/s | 41.3535 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 0.3339ms | 16.5008μs | 60.6030 KOps/s | 63.6363 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 45.6550μs | 13.7953μs | 72.4885 KOps/s | 71.8331 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 43.3410μs | 9.0510μs | 110.4855 KOps/s | 112.5443 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 54.0210μs | 25.3266μs | 39.4842 KOps/s | 35.9447 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 42.2300μs | 16.9444μs | 59.0167 KOps/s | 60.0104 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 53.7810μs | 14.8360μs | 67.4036 KOps/s | 66.5877 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 36.9290μs | 10.2177μs | 97.8698 KOps/s | 100.4916 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 0.2162ms | 24.3102μs | 41.1349 KOps/s | 41.4302 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 43.6420μs | 15.8737μs | 62.9973 KOps/s | 64.6521 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 54.6230μs | 16.1171μs | 62.0459 KOps/s | 61.8539 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 35.0550μs | 10.3125μs | 96.9693 KOps/s | 100.2814 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 40.0750μs | 25.9086μs | 38.5972 KOps/s | 39.2623 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 51.0260μs | 16.9836μs | 58.8802 KOps/s | 60.2053 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 70.2720μs | 17.2500μs | 57.9711 KOps/s | 57.3738 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 56.6670μs | 11.4918μs | 87.0185 KOps/s | 89.7415 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 61.0350μs | 26.7652μs | 37.3620 KOps/s | 37.8383 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 43.9430μs | 18.5397μs | 53.9382 KOps/s | 55.7717 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 55.3540μs | 17.2529μs | 57.9614 KOps/s | 58.3909 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 32.9320μs | 11.4367μs | 87.4381 KOps/s | 90.1659 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 56.9670μs | 27.7088μs | 36.0896 KOps/s | 36.4404 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 80.1660μs | 19.3769μs | 51.6078 KOps/s | 52.9095 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 55.9350μs | 18.2567μs | 54.7744 KOps/s | 55.3863 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 0.1798ms | 13.2461μs | 75.4942 KOps/s | 82.6828 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 10.6917ms | 9.3802ms | 106.6079 Ops/s | 105.9847 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 38.5150ms | 35.7666ms | 27.9590 Ops/s | 29.7554 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.2174ms | 0.1767ms | 5.6588 KOps/s | 5.3537 KOps/s | |
test_values[td1_return_estimate-False-False] | 25.0383ms | 23.3720ms | 42.7862 Ops/s | 42.9779 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 37.1661ms | 36.0361ms | 27.7499 Ops/s | 29.6441 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 34.2336ms | 33.8210ms | 29.5674 Ops/s | 29.8889 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 36.9717ms | 35.8649ms | 27.8824 Ops/s | 29.7563 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 8.2071ms | 8.0806ms | 123.7529 Ops/s | 124.1508 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.3791ms | 1.7909ms | 558.3762 Ops/s | 523.8667 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.4372ms | 0.3538ms | 2.8261 KOps/s | 2.8904 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 48.2365ms | 45.3647ms | 22.0436 Ops/s | 24.7494 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.6168ms | 3.0405ms | 328.8910 Ops/s | 326.5061 Ops/s | |
test_dqn_speed | 7.1471ms | 1.3723ms | 728.7082 Ops/s | 736.5421 Ops/s | |
test_ddpg_speed | 3.1373ms | 2.6937ms | 371.2306 Ops/s | 372.6962 Ops/s | |
test_sac_speed | 9.8827ms | 8.5055ms | 117.5714 Ops/s | 118.9879 Ops/s | |
test_redq_speed | 14.3801ms | 13.4286ms | 74.4679 Ops/s | 74.8018 Ops/s | |
test_redq_deprec_speed | 16.1078ms | 13.9382ms | 71.7455 Ops/s | 72.2565 Ops/s | |
test_td3_speed | 16.6821ms | 8.5488ms | 116.9758 Ops/s | 119.7827 Ops/s | |
test_cql_speed | 40.7966ms | 37.5262ms | 26.6480 Ops/s | 27.2648 Ops/s | |
test_a2c_speed | 8.3583ms | 7.5856ms | 131.8279 Ops/s | 131.4129 Ops/s | |
test_ppo_speed | 9.5450ms | 7.9531ms | 125.7374 Ops/s | 123.0788 Ops/s | |
test_reinforce_speed | 7.7859ms | 6.6853ms | 149.5830 Ops/s | 148.5973 Ops/s | |
test_iql_speed | 34.7283ms | 33.2457ms | 30.0790 Ops/s | 30.2577 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.6450ms | 2.4159ms | 413.9233 Ops/s | 422.7347 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.1983ms | 0.5168ms | 1.9351 KOps/s | 1.9593 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7061ms | 0.4893ms | 2.0437 KOps/s | 2.0676 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.6549ms | 2.4232ms | 412.6848 Ops/s | 403.6833 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6764ms | 0.5044ms | 1.9826 KOps/s | 1.9924 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7605ms | 0.4854ms | 2.0600 KOps/s | 2.0873 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.9177ms | 1.2917ms | 774.1482 Ops/s | 784.9162 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 4.3052ms | 1.2378ms | 807.8959 Ops/s | 829.1202 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 2.6843ms | 2.5479ms | 392.4749 Ops/s | 394.7346 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.1038s | 0.7143ms | 1.4000 KOps/s | 1.6127 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8159ms | 0.5998ms | 1.6672 KOps/s | 1.6910 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.6207ms | 2.4524ms | 407.7564 Ops/s | 417.8342 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.1335ms | 0.5134ms | 1.9480 KOps/s | 1.9526 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7208ms | 0.4908ms | 2.0377 KOps/s | 2.0755 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 2.6681ms | 2.4579ms | 406.8479 Ops/s | 409.4032 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7997ms | 0.5086ms | 1.9661 KOps/s | 1.8573 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 3.7848ms | 0.4927ms | 2.0298 KOps/s | 2.1124 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 2.6835ms | 2.5289ms | 395.4353 Ops/s | 399.2756 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.8319ms | 0.6293ms | 1.5890 KOps/s | 1.6180 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7595ms | 0.6054ms | 1.6517 KOps/s | 1.6885 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1142s | 7.8103ms | 128.0358 Ops/s | 127.9662 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 14.4494ms | 12.0217ms | 83.1828 Ops/s | 80.8320 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 1.0830ms | 1.0319ms | 969.0953 Ops/s | 971.0461 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 89.6332ms | 5.4719ms | 182.7511 Ops/s | 133.4040 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 0.1049s | 13.9155ms | 71.8622 Ops/s | 80.3269 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 1.6232ms | 1.0756ms | 929.7317 Ops/s | 970.3675 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 98.6543ms | 5.9883ms | 166.9922 Ops/s | 173.5516 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 15.3869ms | 12.4988ms | 80.0074 Ops/s | 73.8204 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.0363ms | 1.4969ms | 668.0437 Ops/s | 723.0660 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 0.1021s | 0.1010s | 9.8987 Ops/s | 9.2015 Ops/s | |
test_sync | 91.6832ms | 88.6655ms | 11.2783 Ops/s | 11.2621 Ops/s | |
test_async | 0.1755s | 88.5473ms | 11.2934 Ops/s | 12.5296 Ops/s | |
test_single_pixels | 0.1817s | 0.1201s | 8.3289 Ops/s | 8.8909 Ops/s | |
test_sync_pixels | 69.5616ms | 68.6141ms | 14.5743 Ops/s | 14.6756 Ops/s | |
test_async_pixels | 0.1218s | 55.4289ms | 18.0411 Ops/s | 17.8425 Ops/s | |
test_simple | 0.7308s | 0.6725s | 1.4871 Ops/s | 1.4736 Ops/s | |
test_transformed | 0.9289s | 0.8747s | 1.1432 Ops/s | 1.1276 Ops/s | |
test_serial | 2.1292s | 2.0724s | 0.4825 Ops/s | 0.4767 Ops/s | |
test_parallel | 1.8986s | 1.8396s | 0.5436 Ops/s | 0.5514 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1006ms | 32.1438μs | 31.1102 KOps/s | 30.4386 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 40.1400μs | 19.2247μs | 52.0164 KOps/s | 51.7677 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 54.2010μs | 18.2359μs | 54.8370 KOps/s | 53.4472 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 28.4600μs | 10.9591μs | 91.2487 KOps/s | 90.1721 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 50.0510μs | 34.1713μs | 29.2643 KOps/s | 28.6650 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 39.8110μs | 21.4447μs | 46.6316 KOps/s | 46.6557 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 46.8910μs | 20.1357μs | 49.6630 KOps/s | 48.5282 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 30.0500μs | 12.8650μs | 77.7301 KOps/s | 77.7320 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 0.1257ms | 35.4873μs | 28.1791 KOps/s | 27.5514 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 41.9400μs | 22.8553μs | 43.7534 KOps/s | 43.1082 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 38.1210μs | 20.0663μs | 49.8348 KOps/s | 48.1898 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 29.5000μs | 12.8332μs | 77.9228 KOps/s | 76.9270 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 66.1110μs | 37.3565μs | 26.7691 KOps/s | 26.0157 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 40.2700μs | 24.9742μs | 40.0413 KOps/s | 40.0609 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 61.0810μs | 21.7298μs | 46.0197 KOps/s | 44.7264 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 32.3110μs | 14.7290μs | 67.8933 KOps/s | 68.2159 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 88.2210μs | 35.9873μs | 27.7876 KOps/s | 27.0998 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 50.7610μs | 23.2210μs | 43.0644 KOps/s | 42.6436 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 50.1210μs | 24.2500μs | 41.2370 KOps/s | 40.9267 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 29.7400μs | 14.8913μs | 67.1532 KOps/s | 67.9235 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 72.9910μs | 37.7143μs | 26.5151 KOps/s | 25.6909 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 48.2110μs | 24.9536μs | 40.0745 KOps/s | 39.6779 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 43.4600μs | 25.6828μs | 38.9366 KOps/s | 37.8991 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 33.8500μs | 16.5663μs | 60.3637 KOps/s | 60.6277 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 72.1210μs | 39.2559μs | 25.4739 KOps/s | 24.8622 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 62.0110μs | 26.9757μs | 37.0704 KOps/s | 36.6530 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 0.1092ms | 25.4975μs | 39.2195 KOps/s | 38.2303 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 34.0610μs | 16.0476μs | 62.3146 KOps/s | 60.3506 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 70.8810μs | 40.5858μs | 24.6392 KOps/s | 24.0452 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 54.9210μs | 28.5845μs | 34.9840 KOps/s | 34.6906 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 47.7610μs | 27.5097μs | 36.3508 KOps/s | 35.7302 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 50.1110μs | 18.1614μs | 55.0619 KOps/s | 54.8860 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 25.5054ms | 25.0716ms | 39.8858 Ops/s | 39.5626 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 81.9500ms | 3.2133ms | 311.2077 Ops/s | 290.7841 Ops/s | |
test_values[td0_return_estimate-False-False] | 92.2520μs | 65.3374μs | 15.3052 KOps/s | 15.6837 KOps/s | |
test_values[td1_return_estimate-False-False] | 54.5513ms | 53.3993ms | 18.7269 Ops/s | 18.8266 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 2.0915ms | 1.7657ms | 566.3399 Ops/s | 565.6777 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 86.3468ms | 84.9924ms | 11.7658 Ops/s | 11.6897 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 2.0582ms | 1.7626ms | 567.3533 Ops/s | 566.4509 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 23.3859ms | 23.2231ms | 43.0606 Ops/s | 42.6086 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 0.8780ms | 0.7020ms | 1.4245 KOps/s | 1.4250 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7287ms | 0.6524ms | 1.5329 KOps/s | 1.5355 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.4787ms | 1.4521ms | 688.6374 Ops/s | 687.0951 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.9399ms | 0.6723ms | 1.4874 KOps/s | 1.4853 KOps/s | |
test_dqn_speed | 2.0378ms | 1.3854ms | 721.7934 Ops/s | 690.8706 Ops/s | |
test_ddpg_speed | 3.2033ms | 2.6669ms | 374.9696 Ops/s | 366.9029 Ops/s | |
test_sac_speed | 8.4061ms | 7.8929ms | 126.6957 Ops/s | 123.1594 Ops/s | |
test_redq_speed | 11.0987ms | 10.1566ms | 98.4578 Ops/s | 98.5716 Ops/s | |
test_redq_deprec_speed | 11.5022ms | 10.9670ms | 91.1827 Ops/s | 90.5001 Ops/s | |
test_td3_speed | 8.1899ms | 7.8519ms | 127.3573 Ops/s | 124.3758 Ops/s | |
test_cql_speed | 0.1035s | 26.8718ms | 37.2137 Ops/s | 39.6008 Ops/s | |
test_a2c_speed | 5.5605ms | 5.3340ms | 187.4766 Ops/s | 182.8003 Ops/s | |
test_ppo_speed | 5.9763ms | 5.7630ms | 173.5204 Ops/s | 171.7704 Ops/s | |
test_reinforce_speed | 5.2608ms | 4.3792ms | 228.3497 Ops/s | 225.4061 Ops/s | |
test_iql_speed | 20.2314ms | 19.0823ms | 52.4045 Ops/s | 51.7169 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.9936ms | 2.8216ms | 354.4122 Ops/s | 353.3018 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.4802ms | 0.5338ms | 1.8734 KOps/s | 1.8517 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7206ms | 0.5146ms | 1.9433 KOps/s | 1.9321 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 2.9469ms | 2.8241ms | 354.0988 Ops/s | 350.6929 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.4524ms | 0.5238ms | 1.9092 KOps/s | 1.8633 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7595ms | 0.5004ms | 1.9985 KOps/s | 1.9395 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.5919ms | 1.4946ms | 669.0653 Ops/s | 641.3216 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5927ms | 1.4247ms | 701.8902 Ops/s | 679.2986 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.0092ms | 2.9392ms | 340.2252 Ops/s | 337.6863 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.2529ms | 0.6509ms | 1.5363 KOps/s | 1.3352 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8752ms | 0.6272ms | 1.5944 KOps/s | 1.5664 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.9034ms | 2.8154ms | 355.1836 Ops/s | 351.3118 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6725ms | 0.5333ms | 1.8751 KOps/s | 1.8450 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 4.4014ms | 0.5164ms | 1.9366 KOps/s | 1.9312 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.0281ms | 2.8470ms | 351.2530 Ops/s | 349.6180 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6032ms | 0.5277ms | 1.8950 KOps/s | 1.8641 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7440ms | 0.5081ms | 1.9680 KOps/s | 1.9430 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.0536ms | 2.9499ms | 338.9922 Ops/s | 332.7934 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.4460ms | 0.6579ms | 1.5200 KOps/s | 1.5054 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8619ms | 0.6317ms | 1.5830 KOps/s | 1.5367 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1055s | 8.5838ms | 116.4986 Ops/s | 111.9548 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 16.4645ms | 14.1672ms | 70.5858 Ops/s | 69.0895 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 6.9793ms | 1.2578ms | 795.0279 Ops/s | 961.6707 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 98.7593ms | 6.5898ms | 151.7507 Ops/s | 150.7730 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 16.2594ms | 14.1159ms | 70.8422 Ops/s | 67.9181 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 6.9210ms | 1.2558ms | 796.3145 Ops/s | 918.6811 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1013s | 8.9061ms | 112.2821 Ops/s | 110.9134 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 16.8878ms | 14.5009ms | 68.9615 Ops/s | 67.4762 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 2.6282ms | 1.4504ms | 689.4564 Ops/s | 721.4221 Ops/s |
Big fan! sota might not mean much to first comers I prefer algorithm-implementations to sota-implementations, but this is just my opinion |
(cherry picked from commit c29c345)
(cherry picked from commit c29c345)
I chose the name implementations over algorithms because algorithms can make people think that this is somewhat the official way we recommend to use the library.
These codes have a bunch of opinionated choices (e.g. in the way the experiments are configured) from the people that have helped coding them, and they sometimes don't use the entirety of the features they could from the lib.
Therefore, I wouldn't want users to think that the way to use the library is to copy-paste these examples.
cc @albertbou92 @BY571 @matteobettini @skandermoalla