Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 14, 2025

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 14, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3190

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

❌ 12 New Failures, 1 Unrelated Failure

As of commit 70b82ac with merge base 3d1748f (image):

NEW FAILURES - The following jobs have failed:

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.

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens mentioned this pull request Oct 18, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens added the enhancement New feature or request label Oct 20, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Oct 20, 2025

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.6883μs 82.6439μs 12.1001 KOps/s 11.8379 KOps/s $\color{#35bf28}+2.22\%$
test_tensor_to_bytestream_speed[torch.save] 0.1437ms 0.1431ms 6.9869 KOps/s 7.0507 KOps/s $\color{#d91a1a}-0.90\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1217s 0.1207s 8.2835 Ops/s 8.1339 Ops/s $\color{#35bf28}+1.84\%$
test_tensor_to_bytestream_speed[numpy] 2.8203μs 2.8112μs 355.7261 KOps/s 355.0445 KOps/s $\color{#35bf28}+0.19\%$
test_tensor_to_bytestream_speed[safetensors] 45.1208μs 44.2957μs 22.5755 KOps/s 22.5187 KOps/s $\color{#35bf28}+0.25\%$
test_simple 0.5700s 0.5619s 1.7796 Ops/s 1.7259 Ops/s $\color{#35bf28}+3.11\%$
test_transformed 1.2475s 1.1552s 0.8657 Ops/s 0.8704 Ops/s $\color{#d91a1a}-0.54\%$
test_serial 1.6976s 1.6908s 0.5914 Ops/s 0.5869 Ops/s $\color{#35bf28}+0.78\%$
test_parallel 1.1153s 1.0953s 0.9130 Ops/s 0.9211 Ops/s $\color{#d91a1a}-0.88\%$
test_step_mdp_speed[True-True-True-True-True] 0.2204ms 45.9020μs 21.7855 KOps/s 21.9414 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[True-True-True-True-False] 2.7121ms 25.5646μs 39.1165 KOps/s 38.1558 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[True-True-True-False-True] 0.1338ms 25.8423μs 38.6962 KOps/s 38.6197 KOps/s $\color{#35bf28}+0.20\%$
test_step_mdp_speed[True-True-True-False-False] 36.7700μs 14.1680μs 70.5817 KOps/s 70.5144 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[True-True-False-True-True] 94.6710μs 48.4122μs 20.6559 KOps/s 20.6982 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[True-True-False-True-False] 68.1710μs 28.5655μs 35.0073 KOps/s 35.2924 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[True-True-False-False-True] 58.6710μs 28.7208μs 34.8179 KOps/s 35.1647 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[True-True-False-False-False] 96.2310μs 17.5387μs 57.0168 KOps/s 58.9067 KOps/s $\color{#d91a1a}-3.21\%$
test_step_mdp_speed[True-False-True-True-True] 79.3400μs 51.5736μs 19.3898 KOps/s 19.1495 KOps/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-False-True-True-False] 65.9510μs 31.0267μs 32.2304 KOps/s 32.1421 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-False-True-False-True] 59.1810μs 28.6798μs 34.8678 KOps/s 34.3356 KOps/s $\color{#35bf28}+1.55\%$
test_step_mdp_speed[True-False-True-False-False] 47.7710μs 17.3096μs 57.7713 KOps/s 58.7866 KOps/s $\color{#d91a1a}-1.73\%$
test_step_mdp_speed[True-False-False-True-True] 91.9200μs 54.7602μs 18.2614 KOps/s 18.1277 KOps/s $\color{#35bf28}+0.74\%$
test_step_mdp_speed[True-False-False-True-False] 90.5410μs 34.3047μs 29.1505 KOps/s 29.6354 KOps/s $\color{#d91a1a}-1.64\%$
test_step_mdp_speed[True-False-False-False-True] 69.9010μs 31.6999μs 31.5459 KOps/s 32.3734 KOps/s $\color{#d91a1a}-2.56\%$
test_step_mdp_speed[True-False-False-False-False] 54.4600μs 19.9188μs 50.2038 KOps/s 51.1093 KOps/s $\color{#d91a1a}-1.77\%$
test_step_mdp_speed[False-True-True-True-True] 0.1048ms 51.9782μs 19.2389 KOps/s 19.3878 KOps/s $\color{#d91a1a}-0.77\%$
test_step_mdp_speed[False-True-True-True-False] 69.0800μs 32.1806μs 31.0746 KOps/s 32.3937 KOps/s $\color{#d91a1a}-4.07\%$
test_step_mdp_speed[False-True-True-False-True] 2.3386ms 33.1286μs 30.1854 KOps/s 30.5041 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[False-True-True-False-False] 51.1600μs 19.0862μs 52.3939 KOps/s 53.6206 KOps/s $\color{#d91a1a}-2.29\%$
test_step_mdp_speed[False-True-False-True-True] 95.8910μs 54.3499μs 18.3993 KOps/s 18.5136 KOps/s $\color{#d91a1a}-0.62\%$
test_step_mdp_speed[False-True-False-True-False] 68.6500μs 33.5649μs 29.7930 KOps/s 29.1961 KOps/s $\color{#35bf28}+2.04\%$
test_step_mdp_speed[False-True-False-False-True] 78.3510μs 35.8399μs 27.9018 KOps/s 28.4840 KOps/s $\color{#d91a1a}-2.04\%$
test_step_mdp_speed[False-True-False-False-False] 55.7910μs 22.0316μs 45.3893 KOps/s 46.7855 KOps/s $\color{#d91a1a}-2.98\%$
test_step_mdp_speed[False-False-True-True-True] 95.8700μs 56.7155μs 17.6318 KOps/s 17.5664 KOps/s $\color{#35bf28}+0.37\%$
test_step_mdp_speed[False-False-True-True-False] 68.5110μs 36.9744μs 27.0457 KOps/s 27.1477 KOps/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[False-False-True-False-True] 75.1300μs 35.2956μs 28.3322 KOps/s 28.1739 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-False-True-False-False] 55.3410μs 21.7782μs 45.9175 KOps/s 46.1447 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[False-False-False-True-True] 0.1334ms 59.1435μs 16.9080 KOps/s 16.9280 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-False-False-True-False] 0.1058ms 39.3993μs 25.3812 KOps/s 25.5655 KOps/s $\color{#d91a1a}-0.72\%$
test_step_mdp_speed[False-False-False-False-True] 79.7400μs 38.0627μs 26.2725 KOps/s 25.8214 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[False-False-False-False-False] 56.7010μs 24.3568μs 41.0563 KOps/s 41.7349 KOps/s $\color{#d91a1a}-1.63\%$
test_values[generalized_advantage_estimate-True-True] 11.3629ms 10.4197ms 95.9720 Ops/s 100.0800 Ops/s $\color{#d91a1a}-4.10\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.6338ms 17.6047ms 56.8031 Ops/s 59.0323 Ops/s $\color{#d91a1a}-3.78\%$
test_values[td0_return_estimate-False-False] 0.2211ms 0.1308ms 7.6466 KOps/s 8.1513 KOps/s $\textbf{\color{#d91a1a}-6.19\%}$
test_values[td1_return_estimate-False-False] 28.0299ms 27.5105ms 36.3498 Ops/s 36.4356 Ops/s $\color{#d91a1a}-0.24\%$
test_values[vec_td1_return_estimate-False-False] 19.0013ms 17.6920ms 56.5227 Ops/s 67.6499 Ops/s $\textbf{\color{#d91a1a}-16.45\%}$
test_values[td_lambda_return_estimate-True-False] 41.4197ms 40.9815ms 24.4012 Ops/s 24.3232 Ops/s $\color{#35bf28}+0.32\%$
test_values[vec_td_lambda_return_estimate-True-False] 17.9618ms 17.6537ms 56.6455 Ops/s 72.9590 Ops/s $\textbf{\color{#d91a1a}-22.36\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.9752ms 8.7889ms 113.7802 Ops/s 114.4180 Ops/s $\color{#d91a1a}-0.56\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8248ms 1.4983ms 667.4181 Ops/s 670.1406 Ops/s $\color{#d91a1a}-0.41\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5001ms 0.4279ms 2.3369 KOps/s 2.3985 KOps/s $\color{#d91a1a}-2.57\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.6087ms 33.9523ms 29.4531 Ops/s 31.9882 Ops/s $\textbf{\color{#d91a1a}-7.93\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8279ms 1.7245ms 579.8624 Ops/s 582.8449 Ops/s $\color{#d91a1a}-0.51\%$
test_dqn_speed[False-None] 6.3766ms 1.4309ms 698.8515 Ops/s 698.3552 Ops/s $\color{#35bf28}+0.07\%$
test_dqn_speed[False-backward] 2.1004ms 1.9590ms 510.4695 Ops/s 514.2518 Ops/s $\color{#d91a1a}-0.74\%$
test_dqn_speed[True-None] 0.7344ms 0.5074ms 1.9708 KOps/s 1.8753 KOps/s $\textbf{\color{#35bf28}+5.09\%}$
test_dqn_speed[True-backward] 1.1079ms 0.9636ms 1.0378 KOps/s 907.9464 Ops/s $\textbf{\color{#35bf28}+14.30\%}$
test_dqn_speed[reduce-overhead-None] 0.9387ms 0.5307ms 1.8843 KOps/s 1.9013 KOps/s $\color{#d91a1a}-0.89\%$
test_dqn_speed[reduce-overhead-backward] 0.9940ms 0.9597ms 1.0420 KOps/s 1.0409 KOps/s $\color{#35bf28}+0.10\%$
test_ddpg_speed[False-None] 3.2744ms 2.9105ms 343.5818 Ops/s 346.2821 Ops/s $\color{#d91a1a}-0.78\%$
test_ddpg_speed[False-backward] 4.2481ms 4.1219ms 242.6061 Ops/s 242.7569 Ops/s $\color{#d91a1a}-0.06\%$
test_ddpg_speed[True-None] 1.7425ms 1.3690ms 730.4741 Ops/s 717.3474 Ops/s $\color{#35bf28}+1.83\%$
test_ddpg_speed[True-backward] 2.4120ms 2.3565ms 424.3615 Ops/s 405.0269 Ops/s $\color{#35bf28}+4.77\%$
test_ddpg_speed[reduce-overhead-None] 1.6380ms 1.3707ms 729.5612 Ops/s 729.5401 Ops/s $+0.00\%$
test_ddpg_speed[reduce-overhead-backward] 2.4565ms 2.3297ms 429.2315 Ops/s 415.9784 Ops/s $\color{#35bf28}+3.19\%$
test_sac_speed[False-None] 8.3773ms 7.9494ms 125.7959 Ops/s 123.9942 Ops/s $\color{#35bf28}+1.45\%$
test_sac_speed[False-backward] 11.7382ms 11.2339ms 89.0161 Ops/s 88.7123 Ops/s $\color{#35bf28}+0.34\%$
test_sac_speed[True-None] 2.4534ms 2.0713ms 482.7817 Ops/s 473.5930 Ops/s $\color{#35bf28}+1.94\%$
test_sac_speed[True-backward] 4.1221ms 4.0128ms 249.2017 Ops/s 237.6864 Ops/s $\color{#35bf28}+4.84\%$
test_sac_speed[reduce-overhead-None] 2.4176ms 2.0868ms 479.1914 Ops/s 471.4835 Ops/s $\color{#35bf28}+1.63\%$
test_sac_speed[reduce-overhead-backward] 5.9518ms 4.4043ms 227.0523 Ops/s 244.9828 Ops/s $\textbf{\color{#d91a1a}-7.32\%}$
test_redq_speed[False-None] 13.4350ms 10.5135ms 95.1159 Ops/s 89.1657 Ops/s $\textbf{\color{#35bf28}+6.67\%}$
test_redq_speed[False-backward] 18.4573ms 17.9351ms 55.7565 Ops/s 54.7554 Ops/s $\color{#35bf28}+1.83\%$
test_redq_speed[True-None] 4.5477ms 4.3010ms 232.5038 Ops/s 227.2977 Ops/s $\color{#35bf28}+2.29\%$
test_redq_speed[True-backward] 12.7001ms 10.2381ms 97.6740 Ops/s 100.5498 Ops/s $\color{#d91a1a}-2.86\%$
test_redq_speed[reduce-overhead-None] 4.8356ms 4.2912ms 233.0335 Ops/s 221.0242 Ops/s $\textbf{\color{#35bf28}+5.43\%}$
test_redq_speed[reduce-overhead-backward] 10.2584ms 9.9285ms 100.7205 Ops/s 102.7095 Ops/s $\color{#d91a1a}-1.94\%$
test_redq_deprec_speed[False-None] 11.7119ms 11.1527ms 89.6647 Ops/s 88.4162 Ops/s $\color{#35bf28}+1.41\%$
test_redq_deprec_speed[False-backward] 16.3296ms 15.9861ms 62.5543 Ops/s 61.0574 Ops/s $\color{#35bf28}+2.45\%$
test_redq_deprec_speed[True-None] 3.8971ms 3.5478ms 281.8671 Ops/s 281.0559 Ops/s $\color{#35bf28}+0.29\%$
test_redq_deprec_speed[True-backward] 7.6758ms 7.4748ms 133.7835 Ops/s 137.3818 Ops/s $\color{#d91a1a}-2.62\%$
test_redq_deprec_speed[reduce-overhead-None] 3.8752ms 3.4737ms 287.8803 Ops/s 262.3511 Ops/s $\textbf{\color{#35bf28}+9.73\%}$
test_redq_deprec_speed[reduce-overhead-backward] 7.7459ms 7.5142ms 133.0817 Ops/s 127.5223 Ops/s $\color{#35bf28}+4.36\%$
test_td3_speed[False-None] 8.2684ms 8.0192ms 124.7013 Ops/s 118.6900 Ops/s $\textbf{\color{#35bf28}+5.06\%}$
test_td3_speed[False-backward] 11.4516ms 10.9336ms 91.4611 Ops/s 90.2885 Ops/s $\color{#35bf28}+1.30\%$
test_td3_speed[True-None] 1.8278ms 1.7920ms 558.0316 Ops/s 565.2802 Ops/s $\color{#d91a1a}-1.28\%$
test_td3_speed[True-backward] 3.6824ms 3.5556ms 281.2444 Ops/s 276.6413 Ops/s $\color{#35bf28}+1.66\%$
test_td3_speed[reduce-overhead-None] 1.7766ms 1.7415ms 574.2100 Ops/s 566.4451 Ops/s $\color{#35bf28}+1.37\%$
test_td3_speed[reduce-overhead-backward] 3.6395ms 3.5280ms 283.4487 Ops/s 275.7480 Ops/s $\color{#35bf28}+2.79\%$
test_cql_speed[False-None] 28.6979ms 26.0956ms 38.3207 Ops/s 38.5476 Ops/s $\color{#d91a1a}-0.59\%$
test_cql_speed[False-backward] 39.3687ms 35.4998ms 28.1691 Ops/s 28.0765 Ops/s $\color{#35bf28}+0.33\%$
test_cql_speed[True-None] 15.3439ms 12.4761ms 80.1532 Ops/s 79.8585 Ops/s $\color{#35bf28}+0.37\%$
test_cql_speed[True-backward] 19.4768ms 18.4475ms 54.2080 Ops/s 54.9461 Ops/s $\color{#d91a1a}-1.34\%$
test_cql_speed[reduce-overhead-None] 15.4969ms 12.4847ms 80.0979 Ops/s 79.9659 Ops/s $\color{#35bf28}+0.17\%$
test_cql_speed[reduce-overhead-backward] 18.8534ms 18.4525ms 54.1931 Ops/s 57.0810 Ops/s $\textbf{\color{#d91a1a}-5.06\%}$
test_a2c_speed[False-None] 5.6604ms 5.4534ms 183.3734 Ops/s 184.9723 Ops/s $\color{#d91a1a}-0.86\%$
test_a2c_speed[False-backward] 12.1802ms 11.8814ms 84.1655 Ops/s 84.1169 Ops/s $\color{#35bf28}+0.06\%$
test_a2c_speed[True-None] 3.9032ms 3.6964ms 270.5358 Ops/s 285.6193 Ops/s $\textbf{\color{#d91a1a}-5.28\%}$
test_a2c_speed[True-backward] 8.7263ms 8.4552ms 118.2707 Ops/s 110.6590 Ops/s $\textbf{\color{#35bf28}+6.88\%}$
test_a2c_speed[reduce-overhead-None] 4.0074ms 3.7007ms 270.2227 Ops/s 270.0261 Ops/s $\color{#35bf28}+0.07\%$
test_a2c_speed[reduce-overhead-backward] 9.0871ms 8.8351ms 113.1845 Ops/s 105.9593 Ops/s $\textbf{\color{#35bf28}+6.82\%}$
test_ppo_speed[False-None] 6.3432ms 5.9415ms 168.3083 Ops/s 168.5427 Ops/s $\color{#d91a1a}-0.14\%$
test_ppo_speed[False-backward] 13.0540ms 12.5788ms 79.4988 Ops/s 80.1079 Ops/s $\color{#d91a1a}-0.76\%$
test_ppo_speed[True-None] 3.9692ms 3.6212ms 276.1517 Ops/s 261.6567 Ops/s $\textbf{\color{#35bf28}+5.54\%}$
test_ppo_speed[True-backward] 8.8564ms 8.4382ms 118.5091 Ops/s 118.0066 Ops/s $\color{#35bf28}+0.43\%$
test_ppo_speed[reduce-overhead-None] 3.8394ms 3.6424ms 274.5422 Ops/s 273.2050 Ops/s $\color{#35bf28}+0.49\%$
test_ppo_speed[reduce-overhead-backward] 10.4477ms 9.0066ms 111.0298 Ops/s 113.8409 Ops/s $\color{#d91a1a}-2.47\%$
test_reinforce_speed[False-None] 5.1302ms 4.6538ms 214.8777 Ops/s 216.7507 Ops/s $\color{#d91a1a}-0.86\%$
test_reinforce_speed[False-backward] 7.8710ms 7.3857ms 135.3975 Ops/s 134.0339 Ops/s $\color{#35bf28}+1.02\%$
test_reinforce_speed[True-None] 3.0266ms 2.8588ms 349.7933 Ops/s 345.9371 Ops/s $\color{#35bf28}+1.11\%$
test_reinforce_speed[True-backward] 7.8430ms 7.6205ms 131.2251 Ops/s 124.9818 Ops/s $\color{#35bf28}+5.00\%$
test_reinforce_speed[reduce-overhead-None] 3.2820ms 2.8255ms 353.9148 Ops/s 350.5493 Ops/s $\color{#35bf28}+0.96\%$
test_reinforce_speed[reduce-overhead-backward] 8.1225ms 7.8522ms 127.3523 Ops/s 125.5658 Ops/s $\color{#35bf28}+1.42\%$
test_iql_speed[False-None] 25.6075ms 20.7291ms 48.2414 Ops/s 49.5532 Ops/s $\color{#d91a1a}-2.65\%$
test_iql_speed[False-backward] 31.2930ms 30.3433ms 32.9562 Ops/s 32.3918 Ops/s $\color{#35bf28}+1.74\%$
test_iql_speed[True-None] 8.8198ms 8.4218ms 118.7389 Ops/s 118.6283 Ops/s $\color{#35bf28}+0.09\%$
test_iql_speed[True-backward] 17.1051ms 16.6109ms 60.2013 Ops/s 60.5288 Ops/s $\color{#d91a1a}-0.54\%$
test_iql_speed[reduce-overhead-None] 8.7449ms 8.4440ms 118.4276 Ops/s 113.5303 Ops/s $\color{#35bf28}+4.31\%$
test_iql_speed[reduce-overhead-backward] 17.7673ms 17.2228ms 58.0626 Ops/s 63.4426 Ops/s $\textbf{\color{#d91a1a}-8.48\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.6364ms 6.2496ms 160.0096 Ops/s 163.4262 Ops/s $\color{#d91a1a}-2.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5373ms 0.2797ms 3.5752 KOps/s 3.0817 KOps/s $\textbf{\color{#35bf28}+16.01\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6573ms 0.2640ms 3.7874 KOps/s 3.1570 KOps/s $\textbf{\color{#35bf28}+19.97\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2318ms 5.9427ms 168.2732 Ops/s 169.5163 Ops/s $\color{#d91a1a}-0.73\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6284ms 0.2809ms 3.5595 KOps/s 3.0319 KOps/s $\textbf{\color{#35bf28}+17.40\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6940ms 0.2842ms 3.5186 KOps/s 3.1406 KOps/s $\textbf{\color{#35bf28}+12.04\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4316ms 1.2562ms 796.0745 Ops/s 736.9547 Ops/s $\textbf{\color{#35bf28}+8.02\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5654ms 1.1753ms 850.8298 Ops/s 785.3651 Ops/s $\textbf{\color{#35bf28}+8.34\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.8635ms 6.3283ms 158.0195 Ops/s 166.1276 Ops/s $\color{#d91a1a}-4.88\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1864ms 0.4820ms 2.0747 KOps/s 2.0863 KOps/s $\color{#d91a1a}-0.55\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8794ms 0.4672ms 2.1406 KOps/s 2.1982 KOps/s $\color{#d91a1a}-2.62\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1390ms 5.9647ms 167.6542 Ops/s 170.3047 Ops/s $\color{#d91a1a}-1.56\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.1483ms 0.3809ms 2.6257 KOps/s 793.4076 Ops/s $\textbf{\color{#35bf28}+230.93\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5751ms 0.3199ms 3.1262 KOps/s 3.1621 KOps/s $\color{#d91a1a}-1.13\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2149ms 5.8959ms 169.6081 Ops/s 168.6021 Ops/s $\color{#35bf28}+0.60\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2685ms 0.3530ms 2.8331 KOps/s 3.0328 KOps/s $\textbf{\color{#d91a1a}-6.58\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5363ms 0.2868ms 3.4868 KOps/s 3.1899 KOps/s $\textbf{\color{#35bf28}+9.31\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3132ms 6.1454ms 162.7227 Ops/s 163.0947 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1499ms 0.4550ms 2.1977 KOps/s 2.2530 KOps/s $\color{#d91a1a}-2.46\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8277ms 0.4130ms 2.4212 KOps/s 2.3527 KOps/s $\color{#35bf28}+2.91\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.6580ms 5.1713ms 193.3766 Ops/s 190.8059 Ops/s $\color{#35bf28}+1.35\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.4534ms 2.3640ms 423.0062 Ops/s 448.0775 Ops/s $\textbf{\color{#d91a1a}-5.60\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.2326ms 1.0718ms 933.0342 Ops/s 841.8984 Ops/s $\textbf{\color{#35bf28}+10.83\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5085s 15.3589ms 65.1087 Ops/s 56.2078 Ops/s $\textbf{\color{#35bf28}+15.84\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.0768ms 1.4096ms 709.3985 Ops/s 480.8743 Ops/s $\textbf{\color{#35bf28}+47.52\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 12.5979ms 1.2773ms 782.9318 Ops/s 871.9775 Ops/s $\textbf{\color{#d91a1a}-10.21\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.2929ms 5.3072ms 188.4228 Ops/s 183.8616 Ops/s $\color{#35bf28}+2.48\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.5058ms 2.1818ms 458.3472 Ops/s 457.1389 Ops/s $\color{#35bf28}+0.26\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.9518ms 1.3666ms 731.7365 Ops/s 727.8694 Ops/s $\color{#35bf28}+0.53\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.3795ms 33.3438ms 29.9906 Ops/s 30.1754 Ops/s $\color{#d91a1a}-0.61\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.8629ms 17.8078ms 56.1553 Ops/s 58.9281 Ops/s $\color{#d91a1a}-4.71\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 37.8194ms 34.7792ms 28.7528 Ops/s 28.9035 Ops/s $\color{#d91a1a}-0.52\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.5545ms 17.9821ms 55.6110 Ops/s 57.8017 Ops/s $\color{#d91a1a}-3.79\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 37.5067ms 36.0092ms 27.7707 Ops/s 27.5039 Ops/s $\color{#35bf28}+0.97\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.7242ms 19.3182ms 51.7645 Ops/s 53.0222 Ops/s $\color{#d91a1a}-2.37\%$

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added 14 commits October 22, 2025 12:31
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
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. enhancement New feature or request

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant