Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Jan 19, 2026

Summary

  • Add requests to the Jumanji CI environment dependencies

The jumanji package now requires requests for the Sokoban generator (jumanji/environments/routing/sokoban/generator.py), but it was not listed in the CI environment configuration. This caused the Jumanji CI to fail with:

ModuleNotFoundError: No module named 'requests'

Test plan

  • CI should pass after this change

@vmoens vmoens added the Environments Adds or modifies an environment wrapper label Jan 19, 2026
@pytorch-bot
Copy link

pytorch-bot bot commented Jan 19, 2026

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3349

Note: Links to docs will display an error until the docs builds have been completed.

❌ 3 New Failures, 1 Pending, 2 Unrelated Failures

As of commit 15e6570 with merge base 6a19394 (image):

NEW FAILURES - The following jobs have 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.

@meta-cla meta-cla 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 Jan 19, 2026
@vmoens vmoens force-pushed the fix-jumanji-ci-requests branch from a74edc0 to ad1aac8 Compare January 19, 2026 18:25
- Add `requests` to the Jumanji CI environment dependencies (required by
  jumanji's Sokoban generator)
- Add thread limiting environment variables to avoid pthread_create
  resource exhaustion failures (OMP_NUM_THREADS=1, MKL_NUM_THREADS=1,
  XLA_FLAGS)
- Skip test_jumanji_batch_unlocked tests that cause LLVM memory
  allocation failures in CI due to resource constraints
@vmoens vmoens force-pushed the fix-jumanji-ci-requests branch from ad1aac8 to 15e6570 Compare January 19, 2026 19:08
@github-actions
Copy link

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 148. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.1192μs 81.4561μs 12.2765 KOps/s 11.8986 KOps/s $\color{#35bf28}+3.18\%$
test_tensor_to_bytestream_speed[torch.save] 0.1404ms 0.1397ms 7.1593 KOps/s 7.0522 KOps/s $\color{#35bf28}+1.52\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1243s 0.1233s 8.1113 Ops/s 7.9506 Ops/s $\color{#35bf28}+2.02\%$
test_tensor_to_bytestream_speed[numpy] 2.5986μs 2.5928μs 385.6776 KOps/s 393.2137 KOps/s $\color{#d91a1a}-1.92\%$
test_tensor_to_bytestream_speed[safetensors] 40.0016μs 38.4347μs 26.0182 KOps/s 25.8391 KOps/s $\color{#35bf28}+0.69\%$
test_simple 0.8003s 0.7942s 1.2591 Ops/s 1.2199 Ops/s $\color{#35bf28}+3.22\%$
test_transformed 1.5353s 1.4409s 0.6940 Ops/s 0.6916 Ops/s $\color{#35bf28}+0.34\%$
test_serial 2.3901s 2.2993s 0.4349 Ops/s 0.4267 Ops/s $\color{#35bf28}+1.92\%$
test_parallel 2.0373s 1.9701s 0.5076 Ops/s 0.5202 Ops/s $\color{#d91a1a}-2.43\%$
test_step_mdp_speed[True-True-True-True-True] 0.2543ms 45.7573μs 21.8544 KOps/s 22.3998 KOps/s $\color{#d91a1a}-2.43\%$
test_step_mdp_speed[True-True-True-True-False] 54.0610μs 25.0366μs 39.9415 KOps/s 38.7620 KOps/s $\color{#35bf28}+3.04\%$
test_step_mdp_speed[True-True-True-False-True] 61.4210μs 25.4366μs 39.3134 KOps/s 40.0813 KOps/s $\color{#d91a1a}-1.92\%$
test_step_mdp_speed[True-True-True-False-False] 37.9710μs 14.0572μs 71.1377 KOps/s 70.5147 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[True-True-False-True-True] 78.6220μs 47.8887μs 20.8817 KOps/s 20.7976 KOps/s $\color{#35bf28}+0.40\%$
test_step_mdp_speed[True-True-False-True-False] 62.5400μs 27.8333μs 35.9282 KOps/s 35.5793 KOps/s $\color{#35bf28}+0.98\%$
test_step_mdp_speed[True-True-False-False-True] 79.8610μs 28.5743μs 34.9965 KOps/s 35.5540 KOps/s $\color{#d91a1a}-1.57\%$
test_step_mdp_speed[True-True-False-False-False] 47.7510μs 16.7188μs 59.8131 KOps/s 59.4737 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[True-False-True-True-True] 96.9210μs 51.3578μs 19.4712 KOps/s 19.3633 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-False-True-True-False] 61.1810μs 30.3503μs 32.9486 KOps/s 31.7726 KOps/s $\color{#35bf28}+3.70\%$
test_step_mdp_speed[True-False-True-False-True] 60.6910μs 28.6622μs 34.8892 KOps/s 35.8913 KOps/s $\color{#d91a1a}-2.79\%$
test_step_mdp_speed[True-False-True-False-False] 43.0800μs 16.8771μs 59.2519 KOps/s 58.9818 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[True-False-False-True-True] 81.5720μs 54.0466μs 18.5025 KOps/s 18.7185 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[True-False-False-True-False] 63.6310μs 32.4383μs 30.8277 KOps/s 29.5177 KOps/s $\color{#35bf28}+4.44\%$
test_step_mdp_speed[True-False-False-False-True] 59.2610μs 31.0118μs 32.2458 KOps/s 32.8247 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[True-False-False-False-False] 46.1010μs 19.4218μs 51.4886 KOps/s 51.1392 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[False-True-True-True-True] 83.9920μs 50.3886μs 19.8458 KOps/s 20.3140 KOps/s $\color{#d91a1a}-2.31\%$
test_step_mdp_speed[False-True-True-True-False] 63.3710μs 30.5598μs 32.7227 KOps/s 32.5276 KOps/s $\color{#35bf28}+0.60\%$
test_step_mdp_speed[False-True-True-False-True] 60.3810μs 32.1179μs 31.1353 KOps/s 31.3465 KOps/s $\color{#d91a1a}-0.67\%$
test_step_mdp_speed[False-True-True-False-False] 50.0210μs 18.2750μs 54.7194 KOps/s 54.5956 KOps/s $\color{#35bf28}+0.23\%$
test_step_mdp_speed[False-True-False-True-True] 2.7556ms 53.7437μs 18.6068 KOps/s 18.8168 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[False-True-False-True-False] 76.7810μs 33.5299μs 29.8241 KOps/s 29.5943 KOps/s $\color{#35bf28}+0.78\%$
test_step_mdp_speed[False-True-False-False-True] 75.2210μs 34.6691μs 28.8441 KOps/s 29.1088 KOps/s $\color{#d91a1a}-0.91\%$
test_step_mdp_speed[False-True-False-False-False] 47.7710μs 21.5531μs 46.3970 KOps/s 47.2987 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[False-False-True-True-True] 87.9620μs 56.4315μs 17.7206 KOps/s 18.0044 KOps/s $\color{#d91a1a}-1.58\%$
test_step_mdp_speed[False-False-True-True-False] 0.1121ms 36.0477μs 27.7410 KOps/s 26.9964 KOps/s $\color{#35bf28}+2.76\%$
test_step_mdp_speed[False-False-True-False-True] 69.2010μs 34.9321μs 28.6270 KOps/s 29.4661 KOps/s $\color{#d91a1a}-2.85\%$
test_step_mdp_speed[False-False-True-False-False] 78.7610μs 20.4440μs 48.9141 KOps/s 47.1694 KOps/s $\color{#35bf28}+3.70\%$
test_step_mdp_speed[False-False-False-True-True] 0.1005ms 57.8537μs 17.2850 KOps/s 17.5360 KOps/s $\color{#d91a1a}-1.43\%$
test_step_mdp_speed[False-False-False-True-False] 71.8210μs 38.5802μs 25.9201 KOps/s 25.5747 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[False-False-False-False-True] 0.1046ms 36.6898μs 27.2555 KOps/s 27.2323 KOps/s $\color{#35bf28}+0.09\%$
test_step_mdp_speed[False-False-False-False-False] 49.4710μs 23.6373μs 42.3060 KOps/s 42.1681 KOps/s $\color{#35bf28}+0.33\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.7508s 0.7485s 1.3360 Ops/s 1.2929 Ops/s $\color{#35bf28}+3.33\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7345s 0.6371s 1.5696 Ops/s 1.5665 Ops/s $\color{#35bf28}+0.20\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7618s 1.6878s 0.5925 Ops/s 0.5942 Ops/s $\color{#d91a1a}-0.29\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5425s 1.4607s 0.6846 Ops/s 0.6845 Ops/s $\color{#35bf28}+0.02\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0133s 1.9379s 0.5160 Ops/s 0.5159 Ops/s $\color{#35bf28}+0.02\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7974s 1.7149s 0.5831 Ops/s 0.5831 Ops/s $+0.00\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8392s 4.7406s 0.2109 Ops/s 0.2102 Ops/s $\color{#35bf28}+0.35\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6008s 4.4552s 0.2245 Ops/s 0.2228 Ops/s $\color{#35bf28}+0.74\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1560s 2.0083s 0.4979 Ops/s 0.5017 Ops/s $\color{#d91a1a}-0.74\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.8192s 1.6982s 0.5889 Ops/s 0.5833 Ops/s $\color{#35bf28}+0.95\%$
test_values[generalized_advantage_estimate-True-True] 20.7811ms 20.2690ms 49.3364 Ops/s 47.9131 Ops/s $\color{#35bf28}+2.97\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1329s 3.5710ms 280.0340 Ops/s 265.5192 Ops/s $\textbf{\color{#35bf28}+5.47\%}$
test_values[td0_return_estimate-False-False] 0.1065ms 81.7690μs 12.2296 KOps/s 11.9806 KOps/s $\color{#35bf28}+2.08\%$
test_values[td1_return_estimate-False-False] 49.2023ms 48.0230ms 20.8233 Ops/s 20.1867 Ops/s $\color{#35bf28}+3.15\%$
test_values[vec_td1_return_estimate-False-False] 1.3101ms 1.0801ms 925.8100 Ops/s 918.5128 Ops/s $\color{#35bf28}+0.79\%$
test_values[td_lambda_return_estimate-True-False] 79.2078ms 78.7091ms 12.7050 Ops/s 12.3450 Ops/s $\color{#35bf28}+2.92\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3250ms 1.0788ms 926.9495 Ops/s 921.1379 Ops/s $\color{#35bf28}+0.63\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 20.7758ms 20.3863ms 49.0527 Ops/s 47.5577 Ops/s $\color{#35bf28}+3.14\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0876ms 0.7552ms 1.3241 KOps/s 1.3090 KOps/s $\color{#35bf28}+1.16\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7371ms 0.6729ms 1.4862 KOps/s 1.4639 KOps/s $\color{#35bf28}+1.52\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5493ms 1.4830ms 674.3010 Ops/s 668.7816 Ops/s $\color{#35bf28}+0.83\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7423ms 0.6889ms 1.4515 KOps/s 1.4257 KOps/s $\color{#35bf28}+1.81\%$
test_dqn_speed[False-None] 1.6238ms 1.5434ms 647.9151 Ops/s 634.6186 Ops/s $\color{#35bf28}+2.10\%$
test_dqn_speed[False-backward] 2.4174ms 2.1868ms 457.2977 Ops/s 453.2475 Ops/s $\color{#35bf28}+0.89\%$
test_dqn_speed[True-None] 0.7290ms 0.5875ms 1.7021 KOps/s 1.7207 KOps/s $\color{#d91a1a}-1.08\%$
test_dqn_speed[True-backward] 1.2647ms 1.2321ms 811.6215 Ops/s 885.1296 Ops/s $\textbf{\color{#d91a1a}-8.30\%}$
test_dqn_speed[reduce-overhead-None] 0.6663ms 0.5918ms 1.6898 KOps/s 1.6065 KOps/s $\textbf{\color{#35bf28}+5.19\%}$
test_ddpg_speed[False-None] 3.5786ms 2.8998ms 344.8545 Ops/s 341.1362 Ops/s $\color{#35bf28}+1.09\%$
test_ddpg_speed[False-backward] 4.6065ms 4.2988ms 232.6228 Ops/s 236.9387 Ops/s $\color{#d91a1a}-1.82\%$
test_ddpg_speed[True-None] 1.4103ms 1.3427ms 744.7733 Ops/s 717.5021 Ops/s $\color{#35bf28}+3.80\%$
test_ddpg_speed[True-backward] 2.6463ms 2.5810ms 387.4499 Ops/s 381.1248 Ops/s $\color{#35bf28}+1.66\%$
test_ddpg_speed[reduce-overhead-None] 1.4587ms 1.3663ms 731.9003 Ops/s 715.6061 Ops/s $\color{#35bf28}+2.28\%$
test_sac_speed[False-None] 8.9141ms 8.4188ms 118.7820 Ops/s 116.9267 Ops/s $\color{#35bf28}+1.59\%$
test_sac_speed[False-backward] 12.3406ms 11.5954ms 86.2413 Ops/s 85.6211 Ops/s $\color{#35bf28}+0.72\%$
test_sac_speed[True-None] 2.1400ms 1.8824ms 531.2259 Ops/s 525.4256 Ops/s $\color{#35bf28}+1.10\%$
test_sac_speed[True-backward] 3.7834ms 3.6810ms 271.6633 Ops/s 265.1827 Ops/s $\color{#35bf28}+2.44\%$
test_sac_speed[reduce-overhead-None] 19.0845ms 10.7987ms 92.6040 Ops/s 93.8421 Ops/s $\color{#d91a1a}-1.32\%$
test_redq_deprec_speed[False-None] 9.8957ms 9.3353ms 107.1205 Ops/s 105.2581 Ops/s $\color{#35bf28}+1.77\%$
test_redq_deprec_speed[False-backward] 13.2220ms 12.7294ms 78.5585 Ops/s 77.5056 Ops/s $\color{#35bf28}+1.36\%$
test_redq_deprec_speed[True-None] 2.8746ms 2.5774ms 387.9941 Ops/s 383.8312 Ops/s $\color{#35bf28}+1.08\%$
test_redq_deprec_speed[True-backward] 4.7643ms 4.4187ms 226.3122 Ops/s 223.1363 Ops/s $\color{#35bf28}+1.42\%$
test_redq_deprec_speed[reduce-overhead-None] 15.7407ms 9.6950ms 103.1461 Ops/s 86.5978 Ops/s $\textbf{\color{#35bf28}+19.11\%}$
test_td3_speed[False-None] 8.3463ms 8.2159ms 121.7151 Ops/s 119.9148 Ops/s $\color{#35bf28}+1.50\%$
test_td3_speed[False-backward] 11.3244ms 10.8277ms 92.3561 Ops/s 90.7211 Ops/s $\color{#35bf28}+1.80\%$
test_td3_speed[True-None] 1.7416ms 1.7116ms 584.2561 Ops/s 573.2557 Ops/s $\color{#35bf28}+1.92\%$
test_td3_speed[True-backward] 4.6909ms 3.4153ms 292.8024 Ops/s 289.8941 Ops/s $\color{#35bf28}+1.00\%$
test_td3_speed[reduce-overhead-None] 55.7519ms 24.4306ms 40.9324 Ops/s 39.7996 Ops/s $\color{#35bf28}+2.85\%$
test_cql_speed[False-None] 17.7756ms 17.2766ms 57.8817 Ops/s 57.0585 Ops/s $\color{#35bf28}+1.44\%$
test_cql_speed[False-backward] 23.7109ms 22.8400ms 43.7828 Ops/s 43.1999 Ops/s $\color{#35bf28}+1.35\%$
test_cql_speed[True-None] 3.5149ms 3.4210ms 292.3131 Ops/s 285.0378 Ops/s $\color{#35bf28}+2.55\%$
test_cql_speed[True-backward] 6.1276ms 5.6542ms 176.8588 Ops/s 170.2018 Ops/s $\color{#35bf28}+3.91\%$
test_cql_speed[reduce-overhead-None] 18.5200ms 11.7066ms 85.4218 Ops/s 84.6601 Ops/s $\color{#35bf28}+0.90\%$
test_a2c_speed[False-None] 3.8616ms 3.2319ms 309.4172 Ops/s 302.7184 Ops/s $\color{#35bf28}+2.21\%$
test_a2c_speed[False-backward] 6.5637ms 6.1001ms 163.9318 Ops/s 153.8480 Ops/s $\textbf{\color{#35bf28}+6.55\%}$
test_a2c_speed[True-None] 1.4552ms 1.3518ms 739.7314 Ops/s 726.7488 Ops/s $\color{#35bf28}+1.79\%$
test_a2c_speed[True-backward] 3.0933ms 3.0276ms 330.2900 Ops/s 325.3884 Ops/s $\color{#35bf28}+1.51\%$
test_a2c_speed[reduce-overhead-None] 1.0153ms 0.9561ms 1.0459 KOps/s 1.0468 KOps/s $\color{#d91a1a}-0.08\%$
test_ppo_speed[False-None] 3.9073ms 3.8025ms 262.9857 Ops/s 257.2303 Ops/s $\color{#35bf28}+2.24\%$
test_ppo_speed[False-backward] 7.4152ms 6.8894ms 145.1504 Ops/s 142.2639 Ops/s $\color{#35bf28}+2.03\%$
test_ppo_speed[True-None] 1.6855ms 1.4441ms 692.4581 Ops/s 693.2125 Ops/s $\color{#d91a1a}-0.11\%$
test_ppo_speed[True-backward] 3.7088ms 3.3862ms 295.3143 Ops/s 305.5656 Ops/s $\color{#d91a1a}-3.35\%$
test_ppo_speed[reduce-overhead-None] 1.1410ms 1.0247ms 975.9011 Ops/s 955.0172 Ops/s $\color{#35bf28}+2.19\%$
test_reinforce_speed[False-None] 2.3809ms 2.2769ms 439.1904 Ops/s 429.2711 Ops/s $\color{#35bf28}+2.31\%$
test_reinforce_speed[False-backward] 3.5439ms 3.2991ms 303.1106 Ops/s 288.2257 Ops/s $\textbf{\color{#35bf28}+5.16\%}$
test_reinforce_speed[True-None] 1.3685ms 1.2770ms 783.0550 Ops/s 753.3550 Ops/s $\color{#35bf28}+3.94\%$
test_reinforce_speed[True-backward] 3.0867ms 3.0197ms 331.1625 Ops/s 312.4195 Ops/s $\textbf{\color{#35bf28}+6.00\%}$
test_reinforce_speed[reduce-overhead-None] 0.4756s 10.1974ms 98.0641 Ops/s 93.8376 Ops/s $\color{#35bf28}+4.50\%$
test_iql_speed[False-None] 10.0188ms 9.4584ms 105.7263 Ops/s 104.0586 Ops/s $\color{#35bf28}+1.60\%$
test_iql_speed[False-backward] 13.5675ms 13.1717ms 75.9203 Ops/s 72.7341 Ops/s $\color{#35bf28}+4.38\%$
test_iql_speed[True-None] 2.3379ms 2.2438ms 445.6670 Ops/s 438.2539 Ops/s $\color{#35bf28}+1.69\%$
test_iql_speed[True-backward] 5.2164ms 4.9166ms 203.3914 Ops/s 194.4281 Ops/s $\color{#35bf28}+4.61\%$
test_iql_speed[reduce-overhead-None] 0.7304s 12.6202ms 79.2383 Ops/s 98.4686 Ops/s $\textbf{\color{#d91a1a}-19.53\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1263ms 6.0277ms 165.8994 Ops/s 163.3862 Ops/s $\color{#35bf28}+1.54\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7451ms 0.3717ms 2.6901 KOps/s 3.3957 KOps/s $\textbf{\color{#d91a1a}-20.78\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5878ms 0.3569ms 2.8015 KOps/s 3.6378 KOps/s $\textbf{\color{#d91a1a}-22.99\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1517ms 5.8532ms 170.8461 Ops/s 171.2384 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7715ms 0.2847ms 3.5121 KOps/s 3.2102 KOps/s $\textbf{\color{#35bf28}+9.40\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5359ms 0.3370ms 2.9673 KOps/s 3.6436 KOps/s $\textbf{\color{#d91a1a}-18.56\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7370ms 1.3237ms 755.4645 Ops/s 750.8848 Ops/s $\color{#35bf28}+0.61\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.7190ms 1.3732ms 728.2467 Ops/s 800.4375 Ops/s $\textbf{\color{#d91a1a}-9.02\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.0263ms 6.1170ms 163.4801 Ops/s 168.1798 Ops/s $\color{#d91a1a}-2.79\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2024ms 0.4445ms 2.2497 KOps/s 2.2191 KOps/s $\color{#35bf28}+1.38\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8449ms 0.4592ms 2.1776 KOps/s 2.3231 KOps/s $\textbf{\color{#d91a1a}-6.26\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9377ms 5.8503ms 170.9319 Ops/s 168.5909 Ops/s $\color{#35bf28}+1.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6426ms 0.3777ms 2.6476 KOps/s 3.0484 KOps/s $\textbf{\color{#d91a1a}-13.15\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6376ms 0.3636ms 2.7504 KOps/s 3.2716 KOps/s $\textbf{\color{#d91a1a}-15.93\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9721ms 5.7789ms 173.0435 Ops/s 172.2530 Ops/s $\color{#35bf28}+0.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6863ms 0.2881ms 3.4707 KOps/s 2.9623 KOps/s $\textbf{\color{#35bf28}+17.16\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5334ms 0.2693ms 3.7137 KOps/s 3.2211 KOps/s $\textbf{\color{#35bf28}+15.29\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2435ms 5.9825ms 167.1548 Ops/s 167.4268 Ops/s $\color{#d91a1a}-0.16\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9048ms 0.4479ms 2.2329 KOps/s 1.9740 KOps/s $\textbf{\color{#35bf28}+13.12\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7718ms 0.5154ms 1.9401 KOps/s 2.2920 KOps/s $\textbf{\color{#d91a1a}-15.35\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.5186ms 5.0661ms 197.3921 Ops/s 48.2152 Ops/s $\textbf{\color{#35bf28}+309.40\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.5016ms 2.2415ms 446.1397 Ops/s 520.6610 Ops/s $\textbf{\color{#d91a1a}-14.31\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.4003ms 1.3254ms 754.4665 Ops/s 871.6511 Ops/s $\textbf{\color{#d91a1a}-13.44\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.6084s 17.1959ms 58.1535 Ops/s 193.8764 Ops/s $\textbf{\color{#d91a1a}-70.00\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.9605ms 1.8028ms 554.6788 Ops/s 539.3680 Ops/s $\color{#35bf28}+2.84\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.3573ms 1.2809ms 780.7107 Ops/s 730.5444 Ops/s $\textbf{\color{#35bf28}+6.87\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.7960ms 5.3025ms 188.5914 Ops/s 189.2794 Ops/s $\color{#d91a1a}-0.36\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 12.9325ms 2.1488ms 465.3859 Ops/s 465.6961 Ops/s $\color{#d91a1a}-0.07\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.2891ms 1.0921ms 915.6905 Ops/s 899.2222 Ops/s $\color{#35bf28}+1.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 36.9752ms 34.9081ms 28.6467 Ops/s 20.0794 Ops/s $\textbf{\color{#35bf28}+42.67\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.5397ms 17.9334ms 55.7619 Ops/s 54.4239 Ops/s $\color{#35bf28}+2.46\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 40.2717ms 36.0964ms 27.7036 Ops/s 27.3757 Ops/s $\color{#35bf28}+1.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.7003ms 18.2413ms 54.8206 Ops/s 53.5312 Ops/s $\color{#35bf28}+2.41\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 39.5850ms 37.8787ms 26.4001 Ops/s 26.1141 Ops/s $\color{#35bf28}+1.09\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.3343ms 19.7562ms 50.6170 Ops/s 49.6344 Ops/s $\color{#35bf28}+1.98\%$

@github-actions
Copy link

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 153. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.1348μs 81.1281μs 12.3262 KOps/s 12.1273 KOps/s $\color{#35bf28}+1.64\%$
test_tensor_to_bytestream_speed[torch.save] 0.1428ms 0.1420ms 7.0418 KOps/s 6.9573 KOps/s $\color{#35bf28}+1.21\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1256s 0.1247s 8.0211 Ops/s 8.0249 Ops/s $\color{#d91a1a}-0.05\%$
test_tensor_to_bytestream_speed[numpy] 2.5140μs 2.5102μs 398.3786 KOps/s 401.9456 KOps/s $\color{#d91a1a}-0.89\%$
test_tensor_to_bytestream_speed[safetensors] 40.0145μs 39.6982μs 25.1901 KOps/s 24.7156 KOps/s $\color{#35bf28}+1.92\%$
test_simple 0.6687s 0.5767s 1.7341 Ops/s 1.7048 Ops/s $\color{#35bf28}+1.72\%$
test_transformed 1.2518s 1.1577s 0.8638 Ops/s 0.8495 Ops/s $\color{#35bf28}+1.68\%$
test_serial 1.8050s 1.7088s 0.5852 Ops/s 0.5766 Ops/s $\color{#35bf28}+1.49\%$
test_parallel 1.2160s 1.1310s 0.8842 Ops/s 0.8699 Ops/s $\color{#35bf28}+1.64\%$
test_step_mdp_speed[True-True-True-True-True] 0.3127ms 44.6543μs 22.3943 KOps/s 22.1549 KOps/s $\color{#35bf28}+1.08\%$
test_step_mdp_speed[True-True-True-True-False] 55.9020μs 25.4316μs 39.3211 KOps/s 39.3824 KOps/s $\color{#d91a1a}-0.16\%$
test_step_mdp_speed[True-True-True-False-True] 59.7830μs 24.9195μs 40.1292 KOps/s 39.6905 KOps/s $\color{#35bf28}+1.11\%$
test_step_mdp_speed[True-True-True-False-False] 46.5720μs 13.9125μs 71.8778 KOps/s 71.7437 KOps/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[True-True-False-True-True] 86.6640μs 47.9798μs 20.8421 KOps/s 20.9193 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-True-False-True-False] 62.2530μs 27.8387μs 35.9212 KOps/s 35.8231 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-True-False-False-True] 83.2940μs 27.9602μs 35.7651 KOps/s 35.6113 KOps/s $\color{#35bf28}+0.43\%$
test_step_mdp_speed[True-True-False-False-False] 51.5420μs 16.6091μs 60.2078 KOps/s 59.8251 KOps/s $\color{#35bf28}+0.64\%$
test_step_mdp_speed[True-False-True-True-True] 87.8740μs 51.0051μs 19.6059 KOps/s 19.6077 KOps/s $-0.01\%$
test_step_mdp_speed[True-False-True-True-False] 67.5030μs 31.0001μs 32.2579 KOps/s 32.4894 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[True-False-True-False-True] 60.9130μs 27.8567μs 35.8980 KOps/s 35.9817 KOps/s $\color{#d91a1a}-0.23\%$
test_step_mdp_speed[True-False-True-False-False] 51.8830μs 16.5981μs 60.2479 KOps/s 59.8981 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[True-False-False-True-True] 86.9150μs 52.9542μs 18.8842 KOps/s 18.7569 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[True-False-False-True-False] 66.5930μs 33.0240μs 30.2810 KOps/s 29.9207 KOps/s $\color{#35bf28}+1.20\%$
test_step_mdp_speed[True-False-False-False-True] 64.7740μs 30.2309μs 33.0788 KOps/s 32.7174 KOps/s $\color{#35bf28}+1.10\%$
test_step_mdp_speed[True-False-False-False-False] 48.3630μs 19.1053μs 52.3414 KOps/s 51.4740 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[False-True-True-True-True] 86.8140μs 50.7640μs 19.6990 KOps/s 19.7286 KOps/s $\color{#d91a1a}-0.15\%$
test_step_mdp_speed[False-True-True-True-False] 70.5730μs 30.6591μs 32.6168 KOps/s 32.3425 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[False-True-True-False-True] 72.5630μs 31.4951μs 31.7509 KOps/s 31.4002 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[False-True-True-False-False] 50.4730μs 18.0970μs 55.2577 KOps/s 54.8003 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-True-False-True-True] 2.8773ms 53.4623μs 18.7048 KOps/s 18.8343 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[False-True-False-True-False] 66.5640μs 33.3880μs 29.9509 KOps/s 30.1162 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-True-False-False-True] 77.5630μs 34.3143μs 29.1423 KOps/s 29.4118 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[False-True-False-False-False] 52.9530μs 20.9292μs 47.7801 KOps/s 47.4257 KOps/s $\color{#35bf28}+0.75\%$
test_step_mdp_speed[False-False-True-True-True] 0.1033ms 56.4359μs 17.7192 KOps/s 17.7951 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[False-False-True-True-False] 68.9040μs 36.0026μs 27.7757 KOps/s 27.8327 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[False-False-True-False-True] 70.6330μs 33.7780μs 29.6051 KOps/s 29.4580 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[False-False-True-False-False] 76.7740μs 20.9252μs 47.7892 KOps/s 48.0131 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[False-False-False-True-True] 0.1027ms 57.6517μs 17.3455 KOps/s 17.3512 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[False-False-False-True-False] 72.8430μs 38.4455μs 26.0109 KOps/s 26.0310 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[False-False-False-False-True] 84.5740μs 37.6978μs 26.5267 KOps/s 27.5756 KOps/s $\color{#d91a1a}-3.80\%$
test_step_mdp_speed[False-False-False-False-False] 55.0820μs 23.3718μs 42.7867 KOps/s 42.8199 KOps/s $\color{#d91a1a}-0.08\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8679s 0.7932s 1.2608 Ops/s 1.2824 Ops/s $\color{#d91a1a}-1.69\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7241s 0.6301s 1.5870 Ops/s 1.5454 Ops/s $\color{#35bf28}+2.69\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7494s 1.6776s 0.5961 Ops/s 0.5885 Ops/s $\color{#35bf28}+1.29\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5268s 1.4563s 0.6867 Ops/s 0.6735 Ops/s $\color{#35bf28}+1.95\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0216s 1.9309s 0.5179 Ops/s 0.5108 Ops/s $\color{#35bf28}+1.39\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7807s 1.7036s 0.5870 Ops/s 0.5762 Ops/s $\color{#35bf28}+1.87\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8122s 4.6542s 0.2149 Ops/s 0.2138 Ops/s $\color{#35bf28}+0.49\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.5592s 4.4541s 0.2245 Ops/s 0.2224 Ops/s $\color{#35bf28}+0.94\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0946s 2.0094s 0.4977 Ops/s 0.4900 Ops/s $\color{#35bf28}+1.56\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7658s 1.6834s 0.5940 Ops/s 0.5819 Ops/s $\color{#35bf28}+2.08\%$
test_values[generalized_advantage_estimate-True-True] 11.0454ms 10.8829ms 91.8875 Ops/s 91.4594 Ops/s $\color{#35bf28}+0.47\%$
test_values[vec_generalized_advantage_estimate-True-True] 20.2681ms 18.1216ms 55.1827 Ops/s 55.3892 Ops/s $\color{#d91a1a}-0.37\%$
test_values[td0_return_estimate-False-False] 0.2411ms 0.1343ms 7.4454 KOps/s 7.5752 KOps/s $\color{#d91a1a}-1.71\%$
test_values[td1_return_estimate-False-False] 31.6812ms 30.0967ms 33.2263 Ops/s 33.2851 Ops/s $\color{#d91a1a}-0.18\%$
test_values[vec_td1_return_estimate-False-False] 18.8345ms 18.2037ms 54.9340 Ops/s 55.4590 Ops/s $\color{#d91a1a}-0.95\%$
test_values[td_lambda_return_estimate-True-False] 46.0941ms 44.7563ms 22.3432 Ops/s 22.3919 Ops/s $\color{#d91a1a}-0.22\%$
test_values[vec_td_lambda_return_estimate-True-False] 20.5240ms 18.3604ms 54.4650 Ops/s 55.5849 Ops/s $\color{#d91a1a}-2.01\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.8590ms 9.7585ms 102.4745 Ops/s 102.4434 Ops/s $\color{#35bf28}+0.03\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.9312ms 1.5172ms 659.1050 Ops/s 635.3141 Ops/s $\color{#35bf28}+3.74\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4827ms 0.4339ms 2.3049 KOps/s 2.3135 KOps/s $\color{#d91a1a}-0.37\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.7108ms 32.4117ms 30.8530 Ops/s 28.5356 Ops/s $\textbf{\color{#35bf28}+8.12\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.0466ms 1.7697ms 565.0760 Ops/s 563.0071 Ops/s $\color{#35bf28}+0.37\%$
test_dqn_speed[False-None] 1.7731ms 1.4279ms 700.3303 Ops/s 698.3533 Ops/s $\color{#35bf28}+0.28\%$
test_dqn_speed[False-backward] 2.0441ms 1.9813ms 504.7215 Ops/s 509.4916 Ops/s $\color{#d91a1a}-0.94\%$
test_dqn_speed[True-None] 0.7050ms 0.5518ms 1.8124 KOps/s 1.7975 KOps/s $\color{#35bf28}+0.83\%$
test_dqn_speed[True-backward] 1.0591ms 1.0237ms 976.8933 Ops/s 816.2625 Ops/s $\textbf{\color{#35bf28}+19.68\%}$
test_dqn_speed[reduce-overhead-None] 0.8829ms 0.5417ms 1.8461 KOps/s 1.7623 KOps/s $\color{#35bf28}+4.75\%$
test_ddpg_speed[False-None] 3.1123ms 2.8974ms 345.1394 Ops/s 344.2975 Ops/s $\color{#35bf28}+0.24\%$
test_ddpg_speed[False-backward] 4.5181ms 4.1573ms 240.5386 Ops/s 239.6448 Ops/s $\color{#35bf28}+0.37\%$
test_ddpg_speed[True-None] 1.7790ms 1.4355ms 696.6263 Ops/s 665.6083 Ops/s $\color{#35bf28}+4.66\%$
test_ddpg_speed[True-backward] 2.5102ms 2.4598ms 406.5304 Ops/s 405.5939 Ops/s $\color{#35bf28}+0.23\%$
test_ddpg_speed[reduce-overhead-None] 1.8163ms 1.4253ms 701.6140 Ops/s 689.7334 Ops/s $\color{#35bf28}+1.72\%$
test_sac_speed[False-None] 8.7386ms 8.1886ms 122.1214 Ops/s 120.1270 Ops/s $\color{#35bf28}+1.66\%$
test_sac_speed[False-backward] 12.0193ms 11.5296ms 86.7334 Ops/s 85.0851 Ops/s $\color{#35bf28}+1.94\%$
test_sac_speed[True-None] 2.6199ms 2.2156ms 451.3495 Ops/s 450.2251 Ops/s $\color{#35bf28}+0.25\%$
test_sac_speed[True-backward] 4.5024ms 4.1630ms 240.2120 Ops/s 227.0166 Ops/s $\textbf{\color{#35bf28}+5.81\%}$
test_sac_speed[reduce-overhead-None] 2.5848ms 2.1946ms 455.6576 Ops/s 445.7875 Ops/s $\color{#35bf28}+2.21\%$
test_redq_speed[False-None] 11.0864ms 10.6720ms 93.7030 Ops/s 93.9950 Ops/s $\color{#d91a1a}-0.31\%$
test_redq_speed[False-backward] 20.6645ms 18.5562ms 53.8904 Ops/s 54.8568 Ops/s $\color{#d91a1a}-1.76\%$
test_redq_speed[True-None] 5.4655ms 4.6165ms 216.6162 Ops/s 211.5869 Ops/s $\color{#35bf28}+2.38\%$
test_redq_speed[True-backward] 10.4269ms 10.0694ms 99.3108 Ops/s 100.0197 Ops/s $\color{#d91a1a}-0.71\%$
test_redq_speed[reduce-overhead-None] 4.7120ms 4.5264ms 220.9259 Ops/s 215.3638 Ops/s $\color{#35bf28}+2.58\%$
test_redq_deprec_speed[False-None] 11.8054ms 11.3187ms 88.3498 Ops/s 86.8483 Ops/s $\color{#35bf28}+1.73\%$
test_redq_deprec_speed[False-backward] 16.6174ms 16.1329ms 61.9852 Ops/s 60.3823 Ops/s $\color{#35bf28}+2.65\%$
test_redq_deprec_speed[True-None] 4.3198ms 3.7990ms 263.2271 Ops/s 262.3261 Ops/s $\color{#35bf28}+0.34\%$
test_redq_deprec_speed[True-backward] 8.2279ms 7.9851ms 125.2327 Ops/s 125.1948 Ops/s $\color{#35bf28}+0.03\%$
test_redq_deprec_speed[reduce-overhead-None] 3.8919ms 3.7591ms 266.0186 Ops/s 266.8042 Ops/s $\color{#d91a1a}-0.29\%$
test_td3_speed[False-None] 9.8445ms 8.1717ms 122.3729 Ops/s 121.3622 Ops/s $\color{#35bf28}+0.83\%$
test_td3_speed[False-backward] 11.5006ms 11.0673ms 90.3565 Ops/s 89.2503 Ops/s $\color{#35bf28}+1.24\%$
test_td3_speed[True-None] 2.0058ms 1.9030ms 525.4778 Ops/s 530.3526 Ops/s $\color{#d91a1a}-0.92\%$
test_td3_speed[True-backward] 5.7774ms 4.2655ms 234.4405 Ops/s 238.9829 Ops/s $\color{#d91a1a}-1.90\%$
test_td3_speed[reduce-overhead-None] 1.9084ms 1.8661ms 535.8764 Ops/s 530.2460 Ops/s $\color{#35bf28}+1.06\%$
test_cql_speed[False-None] 29.6016ms 26.6209ms 37.5645 Ops/s 37.6857 Ops/s $\color{#d91a1a}-0.32\%$
test_cql_speed[False-backward] 38.5125ms 36.2621ms 27.5770 Ops/s 27.5847 Ops/s $\color{#d91a1a}-0.03\%$
test_cql_speed[True-None] 13.2436ms 12.7904ms 78.1835 Ops/s 79.9468 Ops/s $\color{#d91a1a}-2.21\%$
test_cql_speed[True-backward] 19.5477ms 18.9246ms 52.8414 Ops/s 54.9367 Ops/s $\color{#d91a1a}-3.81\%$
test_cql_speed[reduce-overhead-None] 13.2392ms 12.7594ms 78.3735 Ops/s 79.2412 Ops/s $\color{#d91a1a}-1.10\%$
test_a2c_speed[False-None] 5.8585ms 5.4582ms 183.2098 Ops/s 183.0873 Ops/s $\color{#35bf28}+0.07\%$
test_a2c_speed[False-backward] 12.5706ms 12.1101ms 82.5757 Ops/s 82.2326 Ops/s $\color{#35bf28}+0.42\%$
test_a2c_speed[True-None] 3.9850ms 3.7822ms 264.3954 Ops/s 261.6081 Ops/s $\color{#35bf28}+1.07\%$
test_a2c_speed[True-backward] 9.0423ms 8.8051ms 113.5711 Ops/s 109.2526 Ops/s $\color{#35bf28}+3.95\%$
test_a2c_speed[reduce-overhead-None] 4.0543ms 3.7846ms 264.2292 Ops/s 263.1812 Ops/s $\color{#35bf28}+0.40\%$
test_ppo_speed[False-None] 6.4012ms 6.1140ms 163.5582 Ops/s 164.6328 Ops/s $\color{#d91a1a}-0.65\%$
test_ppo_speed[False-backward] 13.4666ms 12.9922ms 76.9690 Ops/s 77.2444 Ops/s $\color{#d91a1a}-0.36\%$
test_ppo_speed[True-None] 3.8050ms 3.7032ms 270.0344 Ops/s 270.1853 Ops/s $\color{#d91a1a}-0.06\%$
test_ppo_speed[True-backward] 8.7979ms 8.6150ms 116.0761 Ops/s 115.6465 Ops/s $\color{#35bf28}+0.37\%$
test_ppo_speed[reduce-overhead-None] 3.7972ms 3.6725ms 272.2947 Ops/s 271.5326 Ops/s $\color{#35bf28}+0.28\%$
test_reinforce_speed[False-None] 4.8962ms 4.6862ms 213.3930 Ops/s 210.7207 Ops/s $\color{#35bf28}+1.27\%$
test_reinforce_speed[False-backward] 7.8837ms 7.5544ms 132.3738 Ops/s 131.6779 Ops/s $\color{#35bf28}+0.53\%$
test_reinforce_speed[True-None] 3.0925ms 2.9140ms 343.1655 Ops/s 327.2420 Ops/s $\color{#35bf28}+4.87\%$
test_reinforce_speed[True-backward] 8.1809ms 7.9748ms 125.3945 Ops/s 114.7600 Ops/s $\textbf{\color{#35bf28}+9.27\%}$
test_reinforce_speed[reduce-overhead-None] 3.0918ms 2.9437ms 339.7036 Ops/s 329.7145 Ops/s $\color{#35bf28}+3.03\%$
test_iql_speed[False-None] 26.8688ms 21.0937ms 47.4075 Ops/s 48.8383 Ops/s $\color{#d91a1a}-2.93\%$
test_iql_speed[False-backward] 36.2029ms 31.2153ms 32.0355 Ops/s 32.2869 Ops/s $\color{#d91a1a}-0.78\%$
test_iql_speed[True-None] 9.0515ms 8.7683ms 114.0472 Ops/s 109.8943 Ops/s $\color{#35bf28}+3.78\%$
test_iql_speed[True-backward] 17.8008ms 17.2550ms 57.9541 Ops/s 56.3949 Ops/s $\color{#35bf28}+2.76\%$
test_iql_speed[reduce-overhead-None] 9.0453ms 8.8204ms 113.3731 Ops/s 113.1614 Ops/s $\color{#35bf28}+0.19\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2495ms 6.1004ms 163.9244 Ops/s 162.5461 Ops/s $\color{#35bf28}+0.85\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.4032ms 0.3176ms 3.1489 KOps/s 2.8998 KOps/s $\textbf{\color{#35bf28}+8.59\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5063ms 0.3344ms 2.9902 KOps/s 3.6165 KOps/s $\textbf{\color{#d91a1a}-17.32\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0018ms 5.7844ms 172.8802 Ops/s 170.4528 Ops/s $\color{#35bf28}+1.42\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8726ms 0.3270ms 3.0581 KOps/s 2.9429 KOps/s $\color{#35bf28}+3.91\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5561ms 0.2974ms 3.3625 KOps/s 3.2461 KOps/s $\color{#35bf28}+3.59\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5291ms 1.3228ms 755.9771 Ops/s 685.8692 Ops/s $\textbf{\color{#35bf28}+10.22\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4882ms 1.2472ms 801.7929 Ops/s 728.8496 Ops/s $\textbf{\color{#35bf28}+10.01\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.0672ms 6.1292ms 163.1531 Ops/s 166.6059 Ops/s $\color{#d91a1a}-2.07\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7921ms 0.4840ms 2.0660 KOps/s 1.8043 KOps/s $\textbf{\color{#35bf28}+14.50\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8776ms 0.4576ms 2.1853 KOps/s 2.1660 KOps/s $\color{#35bf28}+0.89\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9496ms 5.8350ms 171.3798 Ops/s 167.3736 Ops/s $\color{#35bf28}+2.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8855ms 0.2901ms 3.4473 KOps/s 2.7399 KOps/s $\textbf{\color{#35bf28}+25.82\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5849ms 0.3517ms 2.8434 KOps/s 3.0180 KOps/s $\textbf{\color{#d91a1a}-5.79\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0026ms 5.7701ms 173.3071 Ops/s 170.8159 Ops/s $\color{#35bf28}+1.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.4857ms 0.3173ms 3.1517 KOps/s 3.3812 KOps/s $\textbf{\color{#d91a1a}-6.79\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 7.1546ms 0.3392ms 2.9481 KOps/s 3.6120 KOps/s $\textbf{\color{#d91a1a}-18.38\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2247ms 5.9411ms 168.3200 Ops/s 164.2992 Ops/s $\color{#35bf28}+2.45\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9166ms 0.5069ms 1.9729 KOps/s 2.1598 KOps/s $\textbf{\color{#d91a1a}-8.66\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8817ms 0.5041ms 1.9838 KOps/s 2.2973 KOps/s $\textbf{\color{#d91a1a}-13.65\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5765s 16.4946ms 60.6261 Ops/s 196.1735 Ops/s $\textbf{\color{#d91a1a}-69.10\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.7866ms 1.7888ms 559.0225 Ops/s 537.3257 Ops/s $\color{#35bf28}+4.04\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.2546ms 0.9085ms 1.1007 KOps/s 1.0851 KOps/s $\color{#35bf28}+1.43\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.1257ms 5.0638ms 197.4794 Ops/s 197.1730 Ops/s $\color{#35bf28}+0.16\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 11.9928ms 1.9356ms 516.6465 Ops/s 488.5950 Ops/s $\textbf{\color{#35bf28}+5.74\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.0340ms 0.9030ms 1.1074 KOps/s 870.1444 Ops/s $\textbf{\color{#35bf28}+27.26\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5408s 15.9982ms 62.5069 Ops/s 58.3695 Ops/s $\textbf{\color{#35bf28}+7.09\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 13.2413ms 2.1829ms 458.1123 Ops/s 505.5021 Ops/s $\textbf{\color{#d91a1a}-9.37\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.7473ms 1.0951ms 913.1994 Ops/s 777.4317 Ops/s $\textbf{\color{#35bf28}+17.46\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.1728ms 35.1789ms 28.4261 Ops/s 28.2605 Ops/s $\color{#35bf28}+0.59\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.1806ms 18.4957ms 54.0667 Ops/s 54.7879 Ops/s $\color{#d91a1a}-1.32\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 39.8110ms 36.1952ms 27.6280 Ops/s 27.3956 Ops/s $\color{#35bf28}+0.85\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.7951ms 18.4084ms 54.3229 Ops/s 53.6830 Ops/s $\color{#35bf28}+1.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 37.9939ms 37.6539ms 26.5577 Ops/s 25.7730 Ops/s $\color{#35bf28}+3.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.9049ms 20.0345ms 49.9140 Ops/s 49.9816 Ops/s $\color{#d91a1a}-0.14\%$

@vmoens vmoens merged commit a4c0e89 into main Jan 19, 2026
104 of 110 checks passed
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. Environments Adds or modifies an environment wrapper

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants