-
Notifications
You must be signed in to change notification settings - Fork 327
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
[Performance] Faster DMC #2002
Merged
Merged
[Performance] Faster DMC #2002
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/2002
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (14 Unrelated Failures)As of commit 8c2d463 with merge base ad73733 (): FLAKY - The following jobs failed but were likely due to flakiness present on trunk:
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
Mar 7, 2024
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 54.9466ms | 54.4435ms | 18.3677 Ops/s | 16.7933 Ops/s | |
test_sync | 38.1087ms | 30.2650ms | 33.0415 Ops/s | 30.2263 Ops/s | |
test_async | 69.6969ms | 28.9454ms | 34.5478 Ops/s | 33.8853 Ops/s | |
test_simple | 0.3996s | 0.3439s | 2.9075 Ops/s | 2.3823 Ops/s | |
test_transformed | 0.4732s | 0.4681s | 2.1364 Ops/s | 1.7738 Ops/s | |
test_serial | 1.2593s | 1.2049s | 0.8299 Ops/s | 0.7755 Ops/s | |
test_parallel | 1.0938s | 1.0445s | 0.9574 Ops/s | 0.8774 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1188ms | 21.0743μs | 47.4511 KOps/s | 46.9723 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 35.1050μs | 12.8739μs | 77.6765 KOps/s | 77.8198 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 50.2150μs | 12.4727μs | 80.1750 KOps/s | 80.1655 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 29.0540μs | 7.4492μs | 134.2420 KOps/s | 133.7036 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 47.6290μs | 22.3032μs | 44.8367 KOps/s | 44.9474 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 32.9010μs | 14.2036μs | 70.4047 KOps/s | 71.0393 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 56.2050μs | 13.6133μs | 73.4575 KOps/s | 74.6957 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 28.5840μs | 8.7657μs | 114.0806 KOps/s | 115.4353 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 51.0950μs | 23.7527μs | 42.1005 KOps/s | 42.2697 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 57.4970μs | 15.4941μs | 64.5405 KOps/s | 65.1902 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 56.7750μs | 13.6544μs | 73.2365 KOps/s | 74.2432 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 29.4940μs | 8.7314μs | 114.5286 KOps/s | 115.5065 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 67.3960μs | 24.8654μs | 40.2165 KOps/s | 40.3204 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 51.7460μs | 16.4867μs | 60.6548 KOps/s | 60.7809 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 61.0240μs | 14.5404μs | 68.7737 KOps/s | 68.4805 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 38.6720μs | 9.8409μs | 101.6166 KOps/s | 102.7851 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 65.2510μs | 23.7407μs | 42.1218 KOps/s | 42.5437 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 88.5550μs | 15.2562μs | 65.5473 KOps/s | 65.5935 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 45.4740μs | 15.7024μs | 63.6846 KOps/s | 63.7918 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 33.9030μs | 10.0019μs | 99.9813 KOps/s | 101.4395 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 46.0260μs | 25.3313μs | 39.4768 KOps/s | 39.4748 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 43.4710μs | 16.6559μs | 60.0388 KOps/s | 60.5008 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 61.3240μs | 16.8992μs | 59.1744 KOps/s | 59.3162 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 31.2380μs | 11.1578μs | 89.6230 KOps/s | 90.1399 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 60.0310μs | 25.9903μs | 38.4759 KOps/s | 38.0415 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 52.7580μs | 17.8736μs | 55.9484 KOps/s | 55.9508 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 39.2030μs | 16.9375μs | 59.0407 KOps/s | 58.7536 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 41.9180μs | 11.0571μs | 90.4397 KOps/s | 92.7597 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 55.8040μs | 27.1391μs | 36.8472 KOps/s | 37.5135 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 65.5500μs | 18.4993μs | 54.0560 KOps/s | 54.0824 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 45.5450μs | 17.8651μs | 55.9750 KOps/s | 56.5182 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 56.9670μs | 12.0616μs | 82.9079 KOps/s | 84.0519 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 12.1555ms | 9.3732ms | 106.6871 Ops/s | 109.3249 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 38.1462ms | 35.6110ms | 28.0812 Ops/s | 28.0127 Ops/s | |
test_values[td0_return_estimate-False-False] | 0.2363ms | 0.1663ms | 6.0123 KOps/s | 6.0827 KOps/s | |
test_values[td1_return_estimate-False-False] | 26.6716ms | 23.1362ms | 43.2224 Ops/s | 43.3607 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 40.6710ms | 35.7374ms | 27.9819 Ops/s | 26.6582 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 36.2816ms | 33.8585ms | 29.5347 Ops/s | 30.1283 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 38.7429ms | 35.6605ms | 28.0422 Ops/s | 27.3061 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 8.3191ms | 8.1582ms | 122.5763 Ops/s | 123.0389 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.6940ms | 2.0757ms | 481.7729 Ops/s | 487.3168 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.5461ms | 0.3549ms | 2.8180 KOps/s | 2.8082 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 52.8779ms | 50.1531ms | 19.9389 Ops/s | 19.7415 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 4.8956ms | 3.0601ms | 326.7827 Ops/s | 329.1803 Ops/s | |
test_dqn_speed | 6.8741ms | 1.3733ms | 728.1971 Ops/s | 740.8297 Ops/s | |
test_ddpg_speed | 3.5466ms | 2.7081ms | 369.2577 Ops/s | 378.2602 Ops/s | |
test_sac_speed | 9.4743ms | 8.2313ms | 121.4869 Ops/s | 122.6106 Ops/s | |
test_redq_speed | 20.5118ms | 13.7747ms | 72.5968 Ops/s | 76.3396 Ops/s | |
test_redq_deprec_speed | 14.0023ms | 13.1660ms | 75.9534 Ops/s | 75.9133 Ops/s | |
test_td3_speed | 8.2969ms | 8.0923ms | 123.5744 Ops/s | 123.9042 Ops/s | |
test_cql_speed | 37.2574ms | 36.3188ms | 27.5340 Ops/s | 27.7421 Ops/s | |
test_a2c_speed | 8.6477ms | 7.3573ms | 135.9191 Ops/s | 136.6136 Ops/s | |
test_ppo_speed | 8.2442ms | 7.6387ms | 130.9117 Ops/s | 132.4777 Ops/s | |
test_reinforce_speed | 7.7244ms | 6.5549ms | 152.5572 Ops/s | 153.4903 Ops/s | |
test_iql_speed | 0.1077s | 35.3098ms | 28.3208 Ops/s | 30.7878 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.6606ms | 2.2101ms | 452.4605 Ops/s | 450.0818 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7530ms | 0.5004ms | 1.9982 KOps/s | 2.0045 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 3.6596ms | 0.4747ms | 2.1064 KOps/s | 2.1097 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 2.4354ms | 2.1559ms | 463.8338 Ops/s | 457.2159 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.8874ms | 0.4913ms | 2.0353 KOps/s | 2.0524 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7624ms | 0.4659ms | 2.1463 KOps/s | 2.1492 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.8124ms | 1.2877ms | 776.5870 Ops/s | 693.9381 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.7023ms | 1.2209ms | 819.0545 Ops/s | 815.7828 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.6463ms | 2.3731ms | 421.3903 Ops/s | 429.2438 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.8506ms | 0.6206ms | 1.6112 KOps/s | 1.6268 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 3.6599ms | 0.5950ms | 1.6806 KOps/s | 1.7054 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 2.5037ms | 2.1886ms | 456.9071 Ops/s | 451.5163 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7968ms | 0.5064ms | 1.9747 KOps/s | 2.0029 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6287ms | 0.4743ms | 2.1082 KOps/s | 1.7182 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 2.6057ms | 2.2387ms | 446.6885 Ops/s | 443.9410 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.0320ms | 0.5004ms | 1.9983 KOps/s | 2.0461 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7147ms | 0.4719ms | 2.1190 KOps/s | 2.1118 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.3295ms | 2.3044ms | 433.9515 Ops/s | 438.7581 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.6951ms | 0.6160ms | 1.6233 KOps/s | 1.3483 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 3.9044ms | 0.6010ms | 1.6639 KOps/s | 1.6952 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1022s | 7.2878ms | 137.2147 Ops/s | 185.6099 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 14.8769ms | 11.9463ms | 83.7077 Ops/s | 83.4100 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 1.0650ms | 1.0222ms | 978.2908 Ops/s | 934.6910 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 91.4867ms | 5.4234ms | 184.3861 Ops/s | 138.0649 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 14.9478ms | 11.9530ms | 83.6607 Ops/s | 83.4027 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 1.5116ms | 1.0239ms | 976.6737 Ops/s | 877.4548 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1008s | 7.6505ms | 130.7110 Ops/s | 174.0527 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 15.1163ms | 12.2787ms | 81.4418 Ops/s | 70.6770 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 1.4740ms | 1.3396ms | 746.4639 Ops/s | 739.7329 Ops/s |
|
Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
---|---|---|---|---|---|
test_single | 0.1053s | 0.1048s | 9.5402 Ops/s | 8.8735 Ops/s | |
test_sync | 93.1055ms | 91.3681ms | 10.9447 Ops/s | 10.5755 Ops/s | |
test_async | 0.1693s | 82.5262ms | 12.1174 Ops/s | 11.0288 Ops/s | |
test_single_pixels | 0.1143s | 0.1136s | 8.8055 Ops/s | 8.2567 Ops/s | |
test_sync_pixels | 69.9057ms | 68.0163ms | 14.7023 Ops/s | 12.3855 Ops/s | |
test_async_pixels | 0.1152s | 56.4100ms | 17.7274 Ops/s | 13.7634 Ops/s | |
test_simple | 0.6885s | 0.6736s | 1.4846 Ops/s | 1.2481 Ops/s | |
test_transformed | 0.8886s | 0.8835s | 1.1319 Ops/s | 0.9849 Ops/s | |
test_serial | 2.2022s | 2.1400s | 0.4673 Ops/s | 0.4552 Ops/s | |
test_parallel | 1.9292s | 1.8662s | 0.5359 Ops/s | 0.5212 Ops/s | |
test_step_mdp_speed[True-True-True-True-True] | 0.1008ms | 33.6243μs | 29.7404 KOps/s | 30.2446 KOps/s | |
test_step_mdp_speed[True-True-True-True-False] | 0.2260ms | 19.8878μs | 50.2821 KOps/s | 51.5465 KOps/s | |
test_step_mdp_speed[True-True-True-False-True] | 0.2137ms | 19.0197μs | 52.5770 KOps/s | 53.5591 KOps/s | |
test_step_mdp_speed[True-True-True-False-False] | 29.1700μs | 11.4359μs | 87.4443 KOps/s | 91.5548 KOps/s | |
test_step_mdp_speed[True-True-False-True-True] | 0.2237ms | 35.2708μs | 28.3521 KOps/s | 29.0685 KOps/s | |
test_step_mdp_speed[True-True-False-True-False] | 0.2094ms | 21.7482μs | 45.9808 KOps/s | 47.0838 KOps/s | |
test_step_mdp_speed[True-True-False-False-True] | 46.3900μs | 20.6178μs | 48.5017 KOps/s | 49.4923 KOps/s | |
test_step_mdp_speed[True-True-False-False-False] | 0.1987ms | 13.4531μs | 74.3323 KOps/s | 77.3809 KOps/s | |
test_step_mdp_speed[True-False-True-True-True] | 53.8800μs | 36.9493μs | 27.0641 KOps/s | 27.2602 KOps/s | |
test_step_mdp_speed[True-False-True-True-False] | 0.2164ms | 23.4900μs | 42.5713 KOps/s | 42.7234 KOps/s | |
test_step_mdp_speed[True-False-True-False-True] | 45.6510μs | 20.6770μs | 48.3630 KOps/s | 50.0540 KOps/s | |
test_step_mdp_speed[True-False-True-False-False] | 0.2021ms | 13.3604μs | 74.8481 KOps/s | 78.4656 KOps/s | |
test_step_mdp_speed[True-False-False-True-True] | 0.2303ms | 38.9099μs | 25.7004 KOps/s | 26.5661 KOps/s | |
test_step_mdp_speed[True-False-False-True-False] | 0.2105ms | 25.4717μs | 39.2592 KOps/s | 40.7500 KOps/s | |
test_step_mdp_speed[True-False-False-False-True] | 37.8200μs | 22.2589μs | 44.9259 KOps/s | 46.6974 KOps/s | |
test_step_mdp_speed[True-False-False-False-False] | 0.2037ms | 15.1446μs | 66.0300 KOps/s | 68.3172 KOps/s | |
test_step_mdp_speed[False-True-True-True-True] | 0.2261ms | 37.3424μs | 26.7792 KOps/s | 27.6234 KOps/s | |
test_step_mdp_speed[False-True-True-True-False] | 0.2130ms | 23.6515μs | 42.2806 KOps/s | 43.3181 KOps/s | |
test_step_mdp_speed[False-True-True-False-True] | 54.7600μs | 25.0185μs | 39.9705 KOps/s | 41.3415 KOps/s | |
test_step_mdp_speed[False-True-True-False-False] | 35.9510μs | 15.0159μs | 66.5961 KOps/s | 67.8640 KOps/s | |
test_step_mdp_speed[False-True-False-True-True] | 0.2367ms | 39.1446μs | 25.5463 KOps/s | 25.9129 KOps/s | |
test_step_mdp_speed[False-True-False-True-False] | 61.4200μs | 25.3885μs | 39.3880 KOps/s | 39.6335 KOps/s | |
test_step_mdp_speed[False-True-False-False-True] | 0.2347ms | 26.1091μs | 38.3008 KOps/s | 39.0226 KOps/s | |
test_step_mdp_speed[False-True-False-False-False] | 0.2251ms | 16.8075μs | 59.4974 KOps/s | 60.4356 KOps/s | |
test_step_mdp_speed[False-False-True-True-True] | 0.2328ms | 40.5414μs | 24.6661 KOps/s | 24.9934 KOps/s | |
test_step_mdp_speed[False-False-True-True-False] | 41.5410μs | 27.4610μs | 36.4153 KOps/s | 37.5841 KOps/s | |
test_step_mdp_speed[False-False-True-False-True] | 0.2141ms | 26.1625μs | 38.2226 KOps/s | 38.5824 KOps/s | |
test_step_mdp_speed[False-False-True-False-False] | 0.2052ms | 16.7848μs | 59.5779 KOps/s | 60.9686 KOps/s | |
test_step_mdp_speed[False-False-False-True-True] | 66.8810μs | 42.0411μs | 23.7862 KOps/s | 24.5155 KOps/s | |
test_step_mdp_speed[False-False-False-True-False] | 50.0800μs | 29.4854μs | 33.9151 KOps/s | 34.7811 KOps/s | |
test_step_mdp_speed[False-False-False-False-True] | 0.2220ms | 28.2840μs | 35.3557 KOps/s | 36.9128 KOps/s | |
test_step_mdp_speed[False-False-False-False-False] | 0.2079ms | 18.9369μs | 52.8070 KOps/s | 55.1709 KOps/s | |
test_values[generalized_advantage_estimate-True-True] | 26.5945ms | 25.8259ms | 38.7208 Ops/s | 40.1434 Ops/s | |
test_values[vec_generalized_advantage_estimate-True-True] | 81.8232ms | 3.1973ms | 312.7599 Ops/s | 298.3300 Ops/s | |
test_values[td0_return_estimate-False-False] | 95.2620μs | 64.4098μs | 15.5256 KOps/s | 15.9943 KOps/s | |
test_values[td1_return_estimate-False-False] | 55.6230ms | 55.2522ms | 18.0988 Ops/s | 18.9888 Ops/s | |
test_values[vec_td1_return_estimate-False-False] | 2.1263ms | 1.7602ms | 568.1048 Ops/s | 572.5939 Ops/s | |
test_values[td_lambda_return_estimate-True-False] | 87.5876ms | 86.8333ms | 11.5163 Ops/s | 11.9753 Ops/s | |
test_values[vec_td_lambda_return_estimate-True-False] | 2.1455ms | 1.7571ms | 569.1277 Ops/s | 572.2012 Ops/s | |
test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.7921ms | 24.2610ms | 41.2184 Ops/s | 42.7412 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 0.8967ms | 0.7011ms | 1.4264 KOps/s | 1.4591 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7145ms | 0.6427ms | 1.5559 KOps/s | 1.5802 KOps/s | |
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.4860ms | 1.4500ms | 689.6318 Ops/s | 695.7410 Ops/s | |
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.9477ms | 0.6610ms | 1.5129 KOps/s | 1.5244 KOps/s | |
test_dqn_speed | 1.6920ms | 1.4638ms | 683.1322 Ops/s | 653.2872 Ops/s | |
test_ddpg_speed | 3.0744ms | 2.7757ms | 360.2713 Ops/s | 377.5191 Ops/s | |
test_sac_speed | 8.6103ms | 8.2382ms | 121.3864 Ops/s | 126.6283 Ops/s | |
test_redq_speed | 11.5923ms | 10.4704ms | 95.5070 Ops/s | 98.5872 Ops/s | |
test_redq_deprec_speed | 11.7309ms | 10.9784ms | 91.0880 Ops/s | 92.0563 Ops/s | |
test_td3_speed | 8.3691ms | 8.1718ms | 122.3714 Ops/s | 128.6381 Ops/s | |
test_cql_speed | 26.7132ms | 25.7107ms | 38.8943 Ops/s | 40.0036 Ops/s | |
test_a2c_speed | 6.3013ms | 5.4951ms | 181.9797 Ops/s | 182.5892 Ops/s | |
test_ppo_speed | 7.4162ms | 5.8398ms | 171.2401 Ops/s | 174.6646 Ops/s | |
test_reinforce_speed | 5.0613ms | 4.4638ms | 224.0232 Ops/s | 223.2150 Ops/s | |
test_iql_speed | 20.1141ms | 19.4622ms | 51.3815 Ops/s | 51.6020 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.1221ms | 2.9141ms | 343.1560 Ops/s | 353.6893 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6560ms | 0.5467ms | 1.8291 KOps/s | 1.8690 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7539ms | 0.5226ms | 1.9134 KOps/s | 1.9471 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.2092ms | 2.9216ms | 342.2724 Ops/s | 352.8219 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7791ms | 0.5423ms | 1.8442 KOps/s | 1.8931 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 4.3877ms | 0.5212ms | 1.9185 KOps/s | 1.9592 KOps/s | |
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.7777ms | 1.5916ms | 628.3162 Ops/s | 657.5156 Ops/s | |
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.7096ms | 1.5161ms | 659.5971 Ops/s | 689.4304 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.2333ms | 3.0524ms | 327.6138 Ops/s | 338.0050 Ops/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.2166ms | 0.6754ms | 1.4805 KOps/s | 1.5070 KOps/s | |
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8891ms | 0.6486ms | 1.5418 KOps/s | 1.5441 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 3.1330ms | 2.9111ms | 343.5168 Ops/s | 352.4896 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6515ms | 0.5475ms | 1.8266 KOps/s | 1.8610 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.1090s | 0.6513ms | 1.5355 KOps/s | 1.9220 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 3.2360ms | 2.9504ms | 338.9392 Ops/s | 347.3809 Ops/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6657ms | 0.5423ms | 1.8441 KOps/s | 1.8798 KOps/s | |
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7055ms | 0.5165ms | 1.9362 KOps/s | 1.9330 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 3.1917ms | 3.0531ms | 327.5371 Ops/s | 337.4994 Ops/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.3305ms | 0.6765ms | 1.4781 KOps/s | 1.4961 KOps/s | |
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.1115s | 0.7899ms | 1.2659 KOps/s | 1.2743 KOps/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1086s | 6.8983ms | 144.9628 Ops/s | 146.6475 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 17.6853ms | 15.4995ms | 64.5183 Ops/s | 67.6092 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 2.3490ms | 1.1469ms | 871.9251 Ops/s | 929.2022 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.1071s | 8.8925ms | 112.4545 Ops/s | 112.8677 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 17.8142ms | 15.4407ms | 64.7639 Ops/s | 68.2595 Ops/s | |
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 2.1314ms | 1.1144ms | 897.3651 Ops/s | 932.5171 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1064s | 7.2378ms | 138.1631 Ops/s | 139.7721 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 18.1205ms | 15.7745ms | 63.3933 Ops/s | 59.2813 Ops/s | |
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 2.5002ms | 1.4632ms | 683.4405 Ops/s | 717.4225 Ops/s |
vmoens
added a commit
that referenced
this pull request
Mar 25, 2024
(cherry picked from commit 358475a)
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.
performance
Performance issue or suggestion for improvement
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.
Other improvements to explore:
to
in env._step may be unnecessary if no mapping is needed_set
could speed things up drasticallycc @teopir