-
Notifications
You must be signed in to change notification settings - Fork 263
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
[NOMERG] Prototyping tensorclass for losses #1892
Open
vmoens
wants to merge
1
commit into
main
Choose a base branch
from
prototype-loss-classes
base: main
Could not load branches
Branch not found: {{ refName }}
Could not load tags
Nothing to show
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
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/1892
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (5 Unrelated Failures)As of commit f949925 with merge base 4d52d5f (): FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
facebook-github-bot
added
the
CLA Signed
This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
label
Feb 9, 2024
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 0.1317s | 66.7379ms | 14.9840 Ops/s | 16.1768 Ops/s | |
test_sync | 43.0471ms | 33.8776ms | 29.5180 Ops/s | 29.7845 Ops/s | |
test_async | 99.9110ms | 33.2546ms | 30.0710 Ops/s | 29.9225 Ops/s | |
test_simple | 0.4970s | 0.4359s | 2.2942 Ops/s | 2.2958 Ops/s | |
test_transformed | 0.6571s | 0.6032s | 1.6578 Ops/s | 1.6726 Ops/s | |
test_serial | 1.5146s | 1.4526s | 0.6884 Ops/s | 0.6947 Ops/s | |
test_parallel | 1.4499s | 1.3902s | 0.7193 Ops/s | 0.7354 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 88.1650μs | 20.9590μs | 47.7123 KOps/s | 47.8348 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 38.3820μs | 12.8344μs | 77.9157 KOps/s | 77.7086 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 34.2040μs | 12.4656μs | 80.2207 KOps/s | 81.7145 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 33.1720μs | 7.5053μs | 133.2396 KOps/s | 132.3081 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 61.9950μs | 22.5489μs | 44.3481 KOps/s | 44.4220 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 37.4800μs | 14.1972μs | 70.4362 KOps/s | 70.8679 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 39.8050μs | 13.6177μs | 73.4337 KOps/s | 73.8477 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 48.3770μs | 8.7709μs | 114.0137 KOps/s | 114.2242 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 67.6060μs | 23.9648μs | 41.7279 KOps/s | 41.9937 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 62.6570μs | 15.5761μs | 64.2011 KOps/s | 64.7681 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 68.4780μs | 13.5192μs | 73.9689 KOps/s | 73.7232 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 32.4310μs | 8.6906μs | 115.0668 KOps/s | 115.0521 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 66.3740μs | 24.5790μs | 40.6851 KOps/s | 39.9482 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 44.7430μs | 16.5635μs | 60.3739 KOps/s | 60.2639 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 62.6070μs | 14.7021μs | 68.0176 KOps/s | 68.7703 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 33.0310μs | 9.9923μs | 100.0770 KOps/s | 100.3671 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 68.3280μs | 23.8000μs | 42.0169 KOps/s | 42.2796 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 60.3930μs | 15.4881μs | 64.5659 KOps/s | 64.7847 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 44.5130μs | 15.7759μs | 63.3880 KOps/s | 63.1991 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 53.1190μs | 9.8077μs | 101.9608 KOps/s | 100.0898 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 50.5440μs | 25.0179μs | 39.9713 KOps/s | 40.0240 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 40.6360μs | 16.6357μs | 60.1119 KOps/s | 60.2332 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 60.5830μs | 16.7080μs | 59.8515 KOps/s | 59.2713 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 59.0500μs | 11.0780μs | 90.2692 KOps/s | 89.9029 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 77.5940μs | 26.2853μs | 38.0441 KOps/s | 38.2894 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 56.4750μs | 17.8522μs | 56.0155 KOps/s | 55.7692 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 43.9720μs | 16.8988μs | 59.1757 KOps/s | 59.2517 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 56.4150μs | 11.0275μs | 90.6827 KOps/s | 89.1656 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 56.7760μs | 27.2094μs | 36.7521 KOps/s | 36.9291 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 0.2886ms | 19.3875μs | 51.5796 KOps/s | 52.7550 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 53.0890μs | 17.9729μs | 55.6395 KOps/s | 55.8695 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 29.7850μs | 12.1660μs | 82.1961 KOps/s | 82.3408 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 11.3920ms | 9.5138ms | 105.1107 Ops/s | 106.6806 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 35.4236ms | 32.9150ms | 30.3813 Ops/s | 28.6882 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.2206ms | 0.1655ms | 6.0421 KOps/s | 6.1572 KOps/s | |
test_values[td1_return_estimate-False-False] | 24.1455ms | 23.8220ms | 41.9780 Ops/s | 44.1222 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 34.2651ms | 33.0600ms | 30.2480 Ops/s | 28.6137 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 37.4565ms | 34.4619ms | 29.0175 Ops/s | 30.2422 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 34.6997ms | 32.8962ms | 30.3987 Ops/s | 28.6020 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 11.3351ms | 8.3587ms | 119.6357 Ops/s | 124.0618 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.6059ms | 1.9797ms | 505.1167 Ops/s | 560.7918 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.4586ms | 0.3536ms | 2.8285 KOps/s | 2.9327 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 40.2049ms | 38.6055ms | 25.9030 Ops/s | 23.5077 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.5874ms | 3.0145ms | 331.7257 Ops/s | 332.1506 Ops/s | |
test_dqn_speed | 1.8557ms | 1.3650ms | 732.5942 Ops/s | 695.3602 Ops/s | |
test_ddpg_speed | 3.1865ms | 2.7272ms | 366.6722 Ops/s | 370.2487 Ops/s | |
test_sac_speed | 11.5429ms | 8.6226ms | 115.9741 Ops/s | 118.4600 Ops/s | |
test_redq_speed | 14.3070ms | 13.2079ms | 75.7124 Ops/s | 76.5414 Ops/s | |
test_redq_deprec_speed | 15.2292ms | 13.4248ms | 74.4892 Ops/s | 75.0865 Ops/s | |
test_td3_speed | 8.8409ms | 8.6440ms | 115.6869 Ops/s | 116.7111 Ops/s | |
test_cql_speed | 37.8557ms | 36.4538ms | 27.4320 Ops/s | 27.4838 Ops/s | |
test_a2c_speed | 8.0616ms | 7.3534ms | 135.9916 Ops/s | 137.2881 Ops/s | |
test_ppo_speed | 8.2274ms | 7.6060ms | 131.4749 Ops/s | 133.0168 Ops/s | |
test_reinforce_speed | 7.5024ms | 6.5784ms | 152.0129 Ops/s | 152.7289 Ops/s | |
test_iql_speed | 34.3591ms | 32.8578ms | 30.4342 Ops/s | 30.5895 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.9634ms | 2.6799ms | 373.1445 Ops/s | 383.2968 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7629ms | 0.5125ms | 1.9512 KOps/s | 1.9734 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.9172ms | 0.4836ms | 2.0677 KOps/s | 2.0737 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.1588ms | 2.6914ms | 371.5541 Ops/s | 370.5536 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6782ms | 0.5027ms | 1.9891 KOps/s | 1.9980 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7713ms | 0.4834ms | 2.0686 KOps/s | 2.0892 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.0846ms | 2.7645ms | 361.7321 Ops/s | 363.7553 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9741ms | 0.6375ms | 1.5686 KOps/s | 1.6139 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 4.9127ms | 0.6106ms | 1.6377 KOps/s | 1.6748 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.9597ms | 2.6952ms | 371.0272 Ops/s | 382.1488 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6113ms | 0.5125ms | 1.9513 KOps/s | 1.9732 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7462ms | 0.4914ms | 2.0350 KOps/s | 2.0659 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.2014ms | 2.7498ms | 363.6631 Ops/s | 348.2003 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6299ms | 0.5099ms | 1.9613 KOps/s | 1.9969 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7138ms | 0.4864ms | 2.0558 KOps/s | 2.0877 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.1028ms | 2.8097ms | 355.9074 Ops/s | 348.5543 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.7653ms | 0.6325ms | 1.5809 KOps/s | 1.6023 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7537ms | 0.6080ms | 1.6448 KOps/s | 1.6738 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1095s | 8.0175ms | 124.7268 Ops/s | 122.9759 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 15.0576ms | 13.1005ms | 76.3330 Ops/s | 77.2690 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 3.7956ms | 2.5195ms | 396.8996 Ops/s | 395.5817 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 92.2882ms | 9.2998ms | 107.5291 Ops/s | 102.1554 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 15.6308ms | 13.1799ms | 75.8731 Ops/s | 72.2723 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 4.0738ms | 2.5459ms | 392.7919 Ops/s | 391.8544 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1067s | 10.0310ms | 99.6913 Ops/s | 125.5193 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 15.5958ms | 13.4472ms | 74.3652 Ops/s | 65.8234 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 4.9960ms | 2.8226ms | 354.2851 Ops/s | 360.6660 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 0.1221s | 0.1177s | 8.4952 Ops/s | 8.3065 Ops/s | |
test_sync | 0.1718s | 0.1030s | 9.7131 Ops/s | 9.7053 Ops/s | |
test_async | 0.2541s | 92.0860ms | 10.8594 Ops/s | 10.9334 Ops/s | |
test_single_pixels | 0.1389s | 0.1380s | 7.2473 Ops/s | 7.2120 Ops/s | |
test_sync_pixels | 83.8074ms | 83.0213ms | 12.0451 Ops/s | 12.4081 Ops/s | |
test_async_pixels | 0.2167s | 74.3765ms | 13.4451 Ops/s | 13.2519 Ops/s | |
test_simple | 0.9064s | 0.8357s | 1.1966 Ops/s | 1.2069 Ops/s | |
test_transformed | 1.1530s | 1.0859s | 0.9209 Ops/s | 0.9107 Ops/s | |
test_serial | 2.5123s | 2.4521s | 0.4078 Ops/s | 0.4093 Ops/s | |
test_parallel | 2.2229s | 2.1228s | 0.4711 Ops/s | 0.4808 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 73.5320μs | 33.1107μs | 30.2017 KOps/s | 30.4348 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 46.5220μs | 19.6602μs | 50.8642 KOps/s | 51.6482 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 53.6510μs | 18.3788μs | 54.4104 KOps/s | 53.7403 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 31.4900μs | 10.8735μs | 91.9665 KOps/s | 89.5362 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 61.0810μs | 33.3664μs | 29.9703 KOps/s | 29.7370 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 45.9110μs | 20.7536μs | 48.1845 KOps/s | 47.5779 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 44.3510μs | 20.0342μs | 49.9147 KOps/s | 49.0423 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 33.3800μs | 12.7732μs | 78.2889 KOps/s | 77.3742 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 64.0510μs | 35.4592μs | 28.2014 KOps/s | 27.9774 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 61.0220μs | 22.5655μs | 44.3154 KOps/s | 43.3115 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 49.3110μs | 20.0256μs | 49.9362 KOps/s | 48.9737 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 31.8210μs | 12.5369μs | 79.7643 KOps/s | 78.9761 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 64.9620μs | 36.7974μs | 27.1758 KOps/s | 26.8348 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 45.3510μs | 24.3945μs | 40.9929 KOps/s | 40.4502 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 47.6620μs | 21.8879μs | 45.6872 KOps/s | 45.7024 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 35.5710μs | 14.3344μs | 69.7622 KOps/s | 68.8340 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 78.1520μs | 35.3893μs | 28.2571 KOps/s | 27.4532 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 47.1410μs | 22.6284μs | 44.1923 KOps/s | 43.3142 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 48.4310μs | 23.7057μs | 42.1840 KOps/s | 41.6068 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 33.6110μs | 14.3729μs | 69.5753 KOps/s | 68.8308 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 70.8320μs | 37.8034μs | 26.4526 KOps/s | 25.9017 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 47.2420μs | 24.5801μs | 40.6834 KOps/s | 40.1889 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 51.3420μs | 25.7088μs | 38.8972 KOps/s | 38.6801 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 36.7810μs | 16.0933μs | 62.1376 KOps/s | 60.2672 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 68.2110μs | 38.9283μs | 25.6882 KOps/s | 25.1311 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 50.2410μs | 26.2073μs | 38.1574 KOps/s | 37.4497 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 46.9010μs | 25.3140μs | 39.5038 KOps/s | 39.0699 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 38.7210μs | 16.0684μs | 62.2339 KOps/s | 61.0442 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 66.6620μs | 40.1442μs | 24.9102 KOps/s | 24.3500 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 55.1110μs | 28.4240μs | 35.1815 KOps/s | 35.2784 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 51.5010μs | 27.4631μs | 36.4125 KOps/s | 37.3028 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 43.7310μs | 18.0333μs | 55.4530 KOps/s | 55.4494 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 27.2212ms | 26.6797ms | 37.4817 Ops/s | 39.1156 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 91.0395ms | 3.3901ms | 294.9806 Ops/s | 294.8425 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.1018ms | 62.2585μs | 16.0621 KOps/s | 16.3261 KOps/s | |
test_values[td1_return_estimate-False-False] | 54.5751ms | 54.1384ms | 18.4712 Ops/s | 18.0452 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 2.0880ms | 1.7672ms | 565.8537 Ops/s | 564.7870 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 92.3128ms | 88.2978ms | 11.3253 Ops/s | 11.5403 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 4.0257ms | 1.8388ms | 543.8289 Ops/s | 552.5542 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.2837ms | 24.0840ms | 41.5214 Ops/s | 39.6430 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 0.8903ms | 0.7085ms | 1.4114 KOps/s | 1.3992 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7241ms | 0.6594ms | 1.5166 KOps/s | 1.5075 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.5451ms | 1.4602ms | 684.8504 Ops/s | 682.1878 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.9416ms | 0.6746ms | 1.4824 KOps/s | 1.4655 KOps/s | |
test_dqn_speed | 8.0148ms | 1.4573ms | 686.2001 Ops/s | 699.1882 Ops/s | |
test_ddpg_speed | 3.2274ms | 2.7714ms | 360.8309 Ops/s | 360.3233 Ops/s | |
test_sac_speed | 9.1391ms | 8.5604ms | 116.8169 Ops/s | 116.7505 Ops/s | |
test_redq_speed | 11.2879ms | 10.6424ms | 93.9636 Ops/s | 94.5424 Ops/s | |
test_redq_deprec_speed | 12.6686ms | 11.8821ms | 84.1600 Ops/s | 82.5704 Ops/s | |
test_td3_speed | 17.6941ms | 8.7953ms | 113.6972 Ops/s | 113.1527 Ops/s | |
test_cql_speed | 27.1657ms | 25.6180ms | 39.0351 Ops/s | 38.8628 Ops/s | |
test_a2c_speed | 6.0731ms | 5.3417ms | 187.2046 Ops/s | 182.4859 Ops/s | |
test_ppo_speed | 6.7748ms | 5.6514ms | 176.9462 Ops/s | 169.8745 Ops/s | |
test_reinforce_speed | 4.8452ms | 4.4293ms | 225.7687 Ops/s | 217.6525 Ops/s | |
test_iql_speed | 20.5029ms | 19.4602ms | 51.3869 Ops/s | 49.9941 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.7086ms | 3.5202ms | 284.0756 Ops/s | 277.8365 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6786ms | 0.5511ms | 1.8146 KOps/s | 1.8013 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6953ms | 0.5264ms | 1.8997 KOps/s | 1.8978 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.6785ms | 3.5491ms | 281.7582 Ops/s | 276.8810 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6743ms | 0.5452ms | 1.8343 KOps/s | 1.8285 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6589ms | 0.5206ms | 1.9207 KOps/s | 1.9172 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.8046ms | 3.6659ms | 272.7873 Ops/s | 266.3998 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.8914ms | 0.6803ms | 1.4699 KOps/s | 1.4708 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8952ms | 0.6522ms | 1.5332 KOps/s | 1.3442 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.7492ms | 3.5399ms | 282.4951 Ops/s | 277.6220 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.8001ms | 0.5525ms | 1.8099 KOps/s | 1.7954 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6657ms | 0.5263ms | 1.9002 KOps/s | 1.8866 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.8989ms | 3.5808ms | 279.2633 Ops/s | 274.4021 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6916ms | 0.5469ms | 1.8286 KOps/s | 1.8169 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7786ms | 0.5212ms | 1.9187 KOps/s | 1.9059 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.8350ms | 3.6848ms | 271.3874 Ops/s | 267.1395 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9254ms | 0.6852ms | 1.4595 KOps/s | 1.4608 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8058ms | 0.6604ms | 1.5143 KOps/s | 1.5226 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1252s | 10.1580ms | 98.4449 Ops/s | 96.3714 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 18.3758ms | 16.0623ms | 62.2577 Ops/s | 61.2868 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 7.1185ms | 3.0545ms | 327.3825 Ops/s | 321.7154 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.1224s | 12.3617ms | 80.8952 Ops/s | 98.3563 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 18.0948ms | 16.0778ms | 62.1975 Ops/s | 61.2134 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 7.3449ms | 3.0351ms | 329.4799 Ops/s | 329.6164 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1231s | 10.4253ms | 95.9202 Ops/s | 78.6963 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 18.7368ms | 16.3837ms | 61.0362 Ops/s | 60.1265 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 8.2807ms | 3.3474ms | 298.7426 Ops/s | 302.3509 Ops/s |
9 tasks
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.
We project on using @Tensorclass to represent losses.
The advantage of tensorclass for losses instead of tensordict is that it will help us use all the features of tensordict while preserving type annotation or even completion.
This draft shows the various steps to integrate this feature:
out_keys
of the loss;Hopefully, all the losses in torchrl.objectives should be patchable with this.