-
Notifications
You must be signed in to change notification settings - Fork 431
[BugFix] Fix CUDA sync in forked subprocess #3269
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
Merged
Merged
Conversation
This file contains hidden or 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/3269
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Dec 31, 2025
This was referenced Dec 31, 2025
|
| Name | Max | Mean | Ops | Ops on Repo HEAD
|
Change |
|---|---|---|---|---|---|
| test_tensor_to_bytestream_speed[pickle] | 82.9600μs | 81.5807μs | 12.2578 KOps/s | 12.2288 KOps/s | |
| test_tensor_to_bytestream_speed[torch.save] | 0.1428ms | 0.1421ms | 7.0349 KOps/s | 6.9617 KOps/s | |
| test_tensor_to_bytestream_speed[untyped_storage] | 0.1275s | 0.1274s | 7.8484 Ops/s | 8.4161 Ops/s | |
| test_tensor_to_bytestream_speed[numpy] | 2.6770μs | 2.6735μs | 374.0427 KOps/s | 374.9000 KOps/s | |
| test_tensor_to_bytestream_speed[safetensors] | 39.5672μs | 39.4447μs | 25.3520 KOps/s | 24.2043 KOps/s | |
| test_simple | 0.5594s | 0.5587s | 1.7900 Ops/s | 1.7184 Ops/s | |
| test_transformed | 1.1410s | 1.1388s | 0.8781 Ops/s | 0.8569 Ops/s | |
| test_serial | 1.6881s | 1.6861s | 0.5931 Ops/s | 0.5832 Ops/s | |
| test_parallel | 1.3003s | 1.1421s | 0.8756 Ops/s | 0.8730 Ops/s | |
| test_step_mdp_speed[True-True-True-True-True] | 0.1552ms | 45.0571μs | 22.1941 KOps/s | 21.9569 KOps/s | |
| test_step_mdp_speed[True-True-True-True-False] | 0.1313ms | 25.8987μs | 38.6119 KOps/s | 39.1170 KOps/s | |
| test_step_mdp_speed[True-True-True-False-True] | 57.2810μs | 25.8343μs | 38.7082 KOps/s | 39.6934 KOps/s | |
| test_step_mdp_speed[True-True-True-False-False] | 37.5610μs | 14.5641μs | 68.6618 KOps/s | 71.2732 KOps/s | |
| test_step_mdp_speed[True-True-False-True-True] | 82.5010μs | 48.9631μs | 20.4235 KOps/s | 20.5314 KOps/s | |
| test_step_mdp_speed[True-True-False-True-False] | 58.0510μs | 28.1946μs | 35.4678 KOps/s | 35.7656 KOps/s | |
| test_step_mdp_speed[True-True-False-False-True] | 61.0510μs | 29.1871μs | 34.2617 KOps/s | 35.1286 KOps/s | |
| test_step_mdp_speed[True-True-False-False-False] | 48.7010μs | 17.4286μs | 57.3769 KOps/s | 58.9057 KOps/s | |
| test_step_mdp_speed[True-False-True-True-True] | 97.9510μs | 51.8873μs | 19.2725 KOps/s | 19.6567 KOps/s | |
| test_step_mdp_speed[True-False-True-True-False] | 63.0210μs | 31.2909μs | 31.9582 KOps/s | 32.1818 KOps/s | |
| test_step_mdp_speed[True-False-True-False-True] | 62.2310μs | 29.0046μs | 34.4773 KOps/s | 35.5036 KOps/s | |
| test_step_mdp_speed[True-False-True-False-False] | 46.8810μs | 17.4381μs | 57.3457 KOps/s | 59.3915 KOps/s | |
| test_step_mdp_speed[True-False-False-True-True] | 97.0410μs | 54.6860μs | 18.2862 KOps/s | 18.9811 KOps/s | |
| test_step_mdp_speed[True-False-False-True-False] | 75.5120μs | 33.8229μs | 29.5658 KOps/s | 29.7005 KOps/s | |
| test_step_mdp_speed[True-False-False-False-True] | 71.5610μs | 31.0670μs | 32.1885 KOps/s | 33.0978 KOps/s | |
| test_step_mdp_speed[True-False-False-False-False] | 48.2210μs | 20.1672μs | 49.5854 KOps/s | 52.0002 KOps/s | |
| test_step_mdp_speed[False-True-True-True-True] | 0.1060ms | 52.5599μs | 19.0259 KOps/s | 19.8532 KOps/s | |
| test_step_mdp_speed[False-True-True-True-False] | 72.8810μs | 31.0777μs | 32.1774 KOps/s | 32.5129 KOps/s | |
| test_step_mdp_speed[False-True-True-False-True] | 2.2937ms | 33.2617μs | 30.0646 KOps/s | 31.3801 KOps/s | |
| test_step_mdp_speed[False-True-True-False-False] | 68.9210μs | 19.2289μs | 52.0051 KOps/s | 54.6675 KOps/s | |
| test_step_mdp_speed[False-True-False-True-True] | 94.5720μs | 54.9755μs | 18.1899 KOps/s | 18.8529 KOps/s | |
| test_step_mdp_speed[False-True-False-True-False] | 66.1710μs | 34.0563μs | 29.3632 KOps/s | 29.7753 KOps/s | |
| test_step_mdp_speed[False-True-False-False-True] | 82.2810μs | 35.3940μs | 28.2534 KOps/s | 29.0646 KOps/s | |
| test_step_mdp_speed[False-True-False-False-False] | 49.9510μs | 21.9885μs | 45.4783 KOps/s | 47.0422 KOps/s | |
| test_step_mdp_speed[False-False-True-True-True] | 0.1325ms | 57.2533μs | 17.4662 KOps/s | 17.7385 KOps/s | |
| test_step_mdp_speed[False-False-True-True-False] | 69.4410μs | 37.2672μs | 26.8332 KOps/s | 27.5512 KOps/s | |
| test_step_mdp_speed[False-False-True-False-True] | 99.0920μs | 35.2114μs | 28.3999 KOps/s | 29.0426 KOps/s | |
| test_step_mdp_speed[False-False-True-False-False] | 50.0910μs | 22.2333μs | 44.9776 KOps/s | 46.7972 KOps/s | |
| test_step_mdp_speed[False-False-False-True-True] | 0.1095ms | 60.0503μs | 16.6527 KOps/s | 17.0182 KOps/s | |
| test_step_mdp_speed[False-False-False-True-False] | 0.1022ms | 39.7170μs | 25.1781 KOps/s | 25.9510 KOps/s | |
| test_step_mdp_speed[False-False-False-False-True] | 70.7710μs | 38.0648μs | 26.2710 KOps/s | 27.2417 KOps/s | |
| test_step_mdp_speed[False-False-False-False-False] | 60.3410μs | 24.6041μs | 40.6436 KOps/s | 42.7082 KOps/s | |
| test_non_tensor_env_rollout_speed[1000-single-True] | 0.8831s | 0.7847s | 1.2744 Ops/s | 1.2707 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-single-False] | 0.7450s | 0.6486s | 1.5417 Ops/s | 1.5349 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] | 1.7921s | 1.7121s | 0.5841 Ops/s | 0.5823 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] | 1.5601s | 1.4821s | 0.6747 Ops/s | 0.6708 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-serial-buffers-True] | 2.0406s | 1.9564s | 0.5112 Ops/s | 0.5096 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-serial-buffers-False] | 1.8145s | 1.7323s | 0.5773 Ops/s | 0.5733 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] | 4.7977s | 4.6777s | 0.2138 Ops/s | 0.2099 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] | 4.6747s | 4.5366s | 0.2204 Ops/s | 0.2219 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] | 2.0704s | 1.9817s | 0.5046 Ops/s | 0.4864 Ops/s | |
| test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] | 1.7959s | 1.7011s | 0.5879 Ops/s | 0.5773 Ops/s | |
| test_values[generalized_advantage_estimate-True-True] | 11.1607ms | 10.6735ms | 93.6902 Ops/s | 96.4921 Ops/s | |
| test_values[vec_generalized_advantage_estimate-True-True] | 19.9009ms | 17.7181ms | 56.4394 Ops/s | 55.8913 Ops/s | |
| test_values[td0_return_estimate-False-False] | 3.6662ms | 0.1997ms | 5.0080 KOps/s | 7.7945 KOps/s | |
| test_values[td1_return_estimate-False-False] | 28.5456ms | 28.2359ms | 35.4159 Ops/s | 35.4681 Ops/s | |
| test_values[vec_td1_return_estimate-False-False] | 21.5598ms | 17.9759ms | 55.6299 Ops/s | 55.8595 Ops/s | |
| test_values[td_lambda_return_estimate-True-False] | 44.6508ms | 41.4898ms | 24.1023 Ops/s | 23.5944 Ops/s | |
| test_values[vec_td_lambda_return_estimate-True-False] | 21.9586ms | 18.0294ms | 55.4648 Ops/s | 55.9345 Ops/s | |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 9.7543ms | 9.5616ms | 104.5854 Ops/s | 107.9471 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.9712ms | 1.5311ms | 653.1179 Ops/s | 663.2621 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.6196ms | 0.4199ms | 2.3816 KOps/s | 2.3421 KOps/s | |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 35.0342ms | 34.4708ms | 29.0101 Ops/s | 28.2850 Ops/s | |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 2.1319ms | 1.8136ms | 551.3991 Ops/s | 543.9403 Ops/s | |
| test_dqn_speed[False-None] | 1.7663ms | 1.4435ms | 692.7565 Ops/s | 706.1464 Ops/s | |
| test_dqn_speed[False-backward] | 2.0994ms | 1.9404ms | 515.3445 Ops/s | 522.1193 Ops/s | |
| test_dqn_speed[True-None] | 0.6934ms | 0.5517ms | 1.8124 KOps/s | 1.8569 KOps/s | |
| test_dqn_speed[True-backward] | 1.1094ms | 0.9730ms | 1.0277 KOps/s | 847.2672 Ops/s | |
| test_dqn_speed[reduce-overhead-None] | 0.9252ms | 0.5331ms | 1.8759 KOps/s | 1.7913 KOps/s | |
| test_dqn_speed[reduce-overhead-backward] | 1.0657ms | 0.9656ms | 1.0356 KOps/s | 899.6286 Ops/s | |
| test_ddpg_speed[False-None] | 3.2556ms | 2.9031ms | 344.4645 Ops/s | 339.8559 Ops/s | |
| test_ddpg_speed[False-backward] | 4.3787ms | 4.1843ms | 238.9868 Ops/s | 239.3225 Ops/s | |
| test_ddpg_speed[True-None] | 1.7718ms | 1.3953ms | 716.6840 Ops/s | 695.8565 Ops/s | |
| test_ddpg_speed[True-backward] | 2.7473ms | 2.3720ms | 421.5940 Ops/s | 371.8922 Ops/s | |
| test_ddpg_speed[reduce-overhead-None] | 1.7757ms | 1.3822ms | 723.4918 Ops/s | 717.8903 Ops/s | |
| test_ddpg_speed[reduce-overhead-backward] | 2.7009ms | 2.3526ms | 425.0700 Ops/s | 360.4926 Ops/s | |
| test_sac_speed[False-None] | 8.4237ms | 7.9676ms | 125.5085 Ops/s | 123.1373 Ops/s | |
| test_sac_speed[False-backward] | 11.6763ms | 11.2574ms | 88.8302 Ops/s | 87.4145 Ops/s | |
| test_sac_speed[True-None] | 3.5921ms | 2.1735ms | 460.0971 Ops/s | 456.3981 Ops/s | |
| test_sac_speed[True-backward] | 4.0993ms | 3.9159ms | 255.3683 Ops/s | 231.9351 Ops/s | |
| test_sac_speed[reduce-overhead-None] | 2.3317ms | 2.1085ms | 474.2697 Ops/s | 465.5475 Ops/s | |
| test_sac_speed[reduce-overhead-backward] | 4.0190ms | 3.9170ms | 255.2998 Ops/s | 215.1847 Ops/s | |
| test_redq_speed[False-None] | 14.9737ms | 10.5349ms | 94.9226 Ops/s | 94.8166 Ops/s | |
| test_redq_speed[False-backward] | 22.8395ms | 17.9775ms | 55.6250 Ops/s | 55.1583 Ops/s | |
| test_redq_speed[True-None] | 4.5257ms | 4.2465ms | 235.4885 Ops/s | 227.2326 Ops/s | |
| test_redq_speed[True-backward] | 9.7155ms | 9.3949ms | 106.4410 Ops/s | 101.1023 Ops/s | |
| test_redq_speed[reduce-overhead-None] | 4.4105ms | 4.2138ms | 237.3149 Ops/s | 227.1550 Ops/s | |
| test_redq_speed[reduce-overhead-backward] | 9.7669ms | 9.5394ms | 104.8283 Ops/s | 103.5275 Ops/s | |
| test_redq_deprec_speed[False-None] | 11.4895ms | 11.0791ms | 90.2602 Ops/s | 88.9044 Ops/s | |
| test_redq_deprec_speed[False-backward] | 16.5447ms | 15.9547ms | 62.6775 Ops/s | 62.6898 Ops/s | |
| test_redq_deprec_speed[True-None] | 4.0511ms | 3.6698ms | 272.4965 Ops/s | 266.1821 Ops/s | |
| test_redq_deprec_speed[True-backward] | 8.0368ms | 7.6533ms | 130.6633 Ops/s | 130.0093 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-None] | 3.9635ms | 3.5887ms | 278.6486 Ops/s | 266.1329 Ops/s | |
| test_redq_deprec_speed[reduce-overhead-backward] | 7.9017ms | 7.6413ms | 130.8685 Ops/s | 114.1697 Ops/s | |
| test_td3_speed[False-None] | 8.1396ms | 8.0394ms | 124.3881 Ops/s | 123.5958 Ops/s | |
| test_td3_speed[False-backward] | 11.4109ms | 10.9294ms | 91.4967 Ops/s | 91.8694 Ops/s | |
| test_td3_speed[True-None] | 1.8475ms | 1.8004ms | 555.4407 Ops/s | 551.1443 Ops/s | |
| test_td3_speed[True-backward] | 3.6936ms | 3.5586ms | 281.0112 Ops/s | 256.1351 Ops/s | |
| test_td3_speed[reduce-overhead-None] | 1.8284ms | 1.7862ms | 559.8604 Ops/s | 561.6152 Ops/s | |
| test_td3_speed[reduce-overhead-backward] | 3.7730ms | 3.6032ms | 277.5321 Ops/s | 274.0959 Ops/s | |
| test_cql_speed[False-None] | 28.5915ms | 25.8926ms | 38.6211 Ops/s | 38.9695 Ops/s | |
| test_cql_speed[False-backward] | 35.8252ms | 34.9792ms | 28.5884 Ops/s | 28.7122 Ops/s | |
| test_cql_speed[True-None] | 12.5478ms | 12.2041ms | 81.9398 Ops/s | 76.9777 Ops/s | |
| test_cql_speed[True-backward] | 18.1602ms | 17.8046ms | 56.1653 Ops/s | 56.0320 Ops/s | |
| test_cql_speed[reduce-overhead-None] | 12.6167ms | 12.2421ms | 81.6855 Ops/s | 80.1235 Ops/s | |
| test_cql_speed[reduce-overhead-backward] | 18.4168ms | 17.9835ms | 55.6065 Ops/s | 56.8270 Ops/s | |
| test_a2c_speed[False-None] | 5.7128ms | 5.3375ms | 187.3546 Ops/s | 183.5946 Ops/s | |
| test_a2c_speed[False-backward] | 12.0893ms | 11.8454ms | 84.4212 Ops/s | 84.2996 Ops/s | |
| test_a2c_speed[True-None] | 3.8118ms | 3.6463ms | 274.2543 Ops/s | 266.9231 Ops/s | |
| test_a2c_speed[True-backward] | 8.7417ms | 8.4859ms | 117.8421 Ops/s | 116.2380 Ops/s | |
| test_a2c_speed[reduce-overhead-None] | 4.0132ms | 3.6912ms | 270.9135 Ops/s | 269.9056 Ops/s | |
| test_a2c_speed[reduce-overhead-backward] | 9.6298ms | 8.6881ms | 115.1004 Ops/s | 111.4911 Ops/s | |
| test_ppo_speed[False-None] | 6.4464ms | 5.8881ms | 169.8346 Ops/s | 169.7147 Ops/s | |
| test_ppo_speed[False-backward] | 12.7435ms | 12.3046ms | 81.2707 Ops/s | 79.9628 Ops/s | |
| test_ppo_speed[True-None] | 4.0168ms | 3.6019ms | 277.6339 Ops/s | 291.7044 Ops/s | |
| test_ppo_speed[True-backward] | 8.6202ms | 8.3227ms | 120.1539 Ops/s | 123.2952 Ops/s | |
| test_ppo_speed[reduce-overhead-None] | 3.9999ms | 3.5386ms | 282.6002 Ops/s | 299.5384 Ops/s | |
| test_ppo_speed[reduce-overhead-backward] | 9.0124ms | 8.5973ms | 116.3152 Ops/s | 119.6108 Ops/s | |
| test_reinforce_speed[False-None] | 4.7466ms | 4.4781ms | 223.3073 Ops/s | 233.8295 Ops/s | |
| test_reinforce_speed[False-backward] | 7.6679ms | 7.3360ms | 136.3148 Ops/s | 140.0910 Ops/s | |
| test_reinforce_speed[True-None] | 3.2749ms | 2.8510ms | 350.7562 Ops/s | 382.9942 Ops/s | |
| test_reinforce_speed[True-backward] | 7.8983ms | 7.5772ms | 131.9744 Ops/s | 130.1005 Ops/s | |
| test_reinforce_speed[reduce-overhead-None] | 3.0795ms | 2.8393ms | 352.1984 Ops/s | 350.9324 Ops/s | |
| test_reinforce_speed[reduce-overhead-backward] | 8.1316ms | 7.8970ms | 126.6303 Ops/s | 120.3207 Ops/s | |
| test_iql_speed[False-None] | 20.3674ms | 19.8489ms | 50.3807 Ops/s | 51.4799 Ops/s | |
| test_iql_speed[False-backward] | 31.1720ms | 30.2813ms | 33.0237 Ops/s | 33.6423 Ops/s | |
| test_iql_speed[True-None] | 8.8187ms | 8.4130ms | 118.8643 Ops/s | 117.3077 Ops/s | |
| test_iql_speed[True-backward] | 16.7584ms | 16.3647ms | 61.1071 Ops/s | 60.9895 Ops/s | |
| test_iql_speed[reduce-overhead-None] | 8.9547ms | 8.4729ms | 118.0234 Ops/s | 117.2019 Ops/s | |
| test_iql_speed[reduce-overhead-backward] | 17.6292ms | 16.7576ms | 59.6744 Ops/s | 59.0478 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 8.1758ms | 6.1898ms | 161.5568 Ops/s | 163.1394 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.5728ms | 0.3358ms | 2.9784 KOps/s | 2.9037 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6387ms | 0.2814ms | 3.5533 KOps/s | 3.0261 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.1983ms | 5.8654ms | 170.4915 Ops/s | 172.9817 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 2.2597ms | 0.3063ms | 3.2643 KOps/s | 2.7937 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5511ms | 0.2849ms | 3.5098 KOps/s | 3.1164 KOps/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.7447ms | 1.2760ms | 783.6889 Ops/s | 789.0783 Ops/s | |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.4471ms | 1.1889ms | 841.1456 Ops/s | 835.0741 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 10.1016ms | 6.1967ms | 161.3768 Ops/s | 165.6673 Ops/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.7992ms | 0.4378ms | 2.2843 KOps/s | 2.1190 KOps/s | |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.6781ms | 0.4402ms | 2.2715 KOps/s | 2.2181 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.0501ms | 5.9110ms | 169.1749 Ops/s | 170.7182 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.6707ms | 0.2856ms | 3.5016 KOps/s | 3.1279 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.5673ms | 0.2639ms | 3.7891 KOps/s | 3.2162 KOps/s | |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.0849ms | 5.8607ms | 170.6271 Ops/s | 172.5847 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.0986ms | 0.2802ms | 3.5694 KOps/s | 639.2559 Ops/s | |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6919ms | 0.2683ms | 3.7271 KOps/s | 3.0691 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.1718ms | 6.0480ms | 165.3440 Ops/s | 163.3198 Ops/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9403ms | 0.4368ms | 2.2896 KOps/s | 2.0712 KOps/s | |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.6953ms | 0.4240ms | 2.3588 KOps/s | 2.2057 KOps/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 6.5873ms | 5.1392ms | 194.5831 Ops/s | 194.7720 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 10.6084ms | 2.3657ms | 422.6999 Ops/s | 484.3612 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 5.7005ms | 1.1428ms | 875.0071 Ops/s | 848.2247 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 6.6764ms | 5.1840ms | 192.9014 Ops/s | 196.4139 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 6.0109ms | 2.3319ms | 428.8378 Ops/s | 476.1597 Ops/s | |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 9.3440ms | 1.3236ms | 755.5270 Ops/s | 914.4292 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.6176s | 17.7101ms | 56.4648 Ops/s | 53.4709 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 9.4124ms | 2.1648ms | 461.9419 Ops/s | 551.4373 Ops/s | |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 1.2886ms | 1.0281ms | 972.6813 Ops/s | 974.6263 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] | 36.0362ms | 34.1641ms | 29.2705 Ops/s | 28.6632 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] | 18.9325ms | 17.5709ms | 56.9123 Ops/s | 55.1847 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] | 37.0964ms | 35.0279ms | 28.5487 Ops/s | 27.0119 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] | 19.4062ms | 17.9276ms | 55.7799 Ops/s | 53.7593 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] | 39.9388ms | 37.4315ms | 26.7155 Ops/s | 25.7287 Ops/s | |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] | 22.6060ms | 19.6284ms | 50.9466 Ops/s | 49.3422 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.
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):