Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 25, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 25, 2025
ghstack-source-id: 19717b2
Pull-Request: #3223
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 25, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3223

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

❌ 1 New Failure, 39 Pending, 4 Unrelated Failures

As of commit 09f9d27 with merge base 3d1748f (image):

NEW FAILURE - The following job has 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 Oct 25, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 25, 2025
ghstack-source-id: a3872d1
Pull-Request: #3223
@github-actions
Copy link

github-actions bot commented Oct 25, 2025

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}18$. 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] 85.3240μs 83.3262μs 12.0010 KOps/s 12.2439 KOps/s $\color{#d91a1a}-1.98\%$
test_tensor_to_bytestream_speed[torch.save] 0.1466ms 0.1451ms 6.8899 KOps/s 7.1266 KOps/s $\color{#d91a1a}-3.32\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1320s 0.1310s 7.6334 Ops/s 8.3534 Ops/s $\textbf{\color{#d91a1a}-8.62\%}$
test_tensor_to_bytestream_speed[numpy] 2.8832μs 2.8787μs 347.3824 KOps/s 360.7186 KOps/s $\color{#d91a1a}-3.70\%$
test_tensor_to_bytestream_speed[safetensors] 44.9394μs 44.7050μs 22.3689 KOps/s 24.0653 KOps/s $\textbf{\color{#d91a1a}-7.05\%}$
test_simple 0.5510s 0.5501s 1.8179 Ops/s 1.7448 Ops/s $\color{#35bf28}+4.19\%$
test_transformed 1.2316s 1.1368s 0.8797 Ops/s 0.8797 Ops/s $-0.01\%$
test_serial 1.6761s 1.6747s 0.5971 Ops/s 0.5937 Ops/s $\color{#35bf28}+0.57\%$
test_parallel 1.1794s 1.0843s 0.9223 Ops/s 0.9438 Ops/s $\color{#d91a1a}-2.28\%$
test_step_mdp_speed[True-True-True-True-True] 0.2257ms 44.1559μs 22.6470 KOps/s 22.7807 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[True-True-True-True-False] 0.1083ms 24.9382μs 40.0992 KOps/s 40.2716 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[True-True-True-False-True] 68.4930μs 24.7927μs 40.3344 KOps/s 40.3568 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[True-True-True-False-False] 94.9550μs 13.7471μs 72.7427 KOps/s 73.0573 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[True-True-False-True-True] 0.1382ms 46.9894μs 21.2814 KOps/s 21.2923 KOps/s $\color{#d91a1a}-0.05\%$
test_step_mdp_speed[True-True-False-True-False] 0.1110ms 27.8681μs 35.8833 KOps/s 36.7080 KOps/s $\color{#d91a1a}-2.25\%$
test_step_mdp_speed[True-True-False-False-True] 70.6130μs 27.6819μs 36.1247 KOps/s 36.9336 KOps/s $\color{#d91a1a}-2.19\%$
test_step_mdp_speed[True-True-False-False-False] 98.3450μs 16.6843μs 59.9366 KOps/s 61.1687 KOps/s $\color{#d91a1a}-2.01\%$
test_step_mdp_speed[True-False-True-True-True] 0.1393ms 50.9189μs 19.6391 KOps/s 20.0053 KOps/s $\color{#d91a1a}-1.83\%$
test_step_mdp_speed[True-False-True-True-False] 82.5040μs 30.4771μs 32.8115 KOps/s 33.2422 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[True-False-True-False-True] 0.1209ms 27.8765μs 35.8725 KOps/s 36.8639 KOps/s $\color{#d91a1a}-2.69\%$
test_step_mdp_speed[True-False-True-False-False] 98.9160μs 16.6510μs 60.0564 KOps/s 61.2211 KOps/s $\color{#d91a1a}-1.90\%$
test_step_mdp_speed[True-False-False-True-True] 0.1329ms 52.6983μs 18.9760 KOps/s 19.2434 KOps/s $\color{#d91a1a}-1.39\%$
test_step_mdp_speed[True-False-False-True-False] 0.1183ms 33.3701μs 29.9669 KOps/s 30.6315 KOps/s $\color{#d91a1a}-2.17\%$
test_step_mdp_speed[True-False-False-False-True] 0.7701ms 30.4523μs 32.8382 KOps/s 33.5426 KOps/s $\color{#d91a1a}-2.10\%$
test_step_mdp_speed[True-False-False-False-False] 48.9630μs 19.2771μs 51.8750 KOps/s 52.4702 KOps/s $\color{#d91a1a}-1.13\%$
test_step_mdp_speed[False-True-True-True-True] 0.1286ms 49.8504μs 20.0600 KOps/s 19.9623 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[False-True-True-True-False] 0.1096ms 30.6408μs 32.6362 KOps/s 33.0064 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[False-True-True-False-True] 2.4673ms 31.8646μs 31.3828 KOps/s 32.2079 KOps/s $\color{#d91a1a}-2.56\%$
test_step_mdp_speed[False-True-True-False-False] 0.1140ms 18.3195μs 54.5868 KOps/s 55.5107 KOps/s $\color{#d91a1a}-1.66\%$
test_step_mdp_speed[False-True-False-True-True] 0.1355ms 52.9426μs 18.8884 KOps/s 18.9996 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[False-True-False-True-False] 0.1180ms 33.1705μs 30.1473 KOps/s 30.1668 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[False-True-False-False-True] 88.0750μs 34.0869μs 29.3368 KOps/s 29.3452 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[False-True-False-False-False] 0.1083ms 20.9599μs 47.7101 KOps/s 47.6987 KOps/s $\color{#35bf28}+0.02\%$
test_step_mdp_speed[False-False-True-True-True] 0.1386ms 54.5378μs 18.3359 KOps/s 18.1980 KOps/s $\color{#35bf28}+0.76\%$
test_step_mdp_speed[False-False-True-True-False] 84.4450μs 36.2185μs 27.6102 KOps/s 27.7925 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[False-False-True-False-True] 0.1135ms 34.1730μs 29.2628 KOps/s 29.3226 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[False-False-True-False-False] 0.1019ms 20.7200μs 48.2626 KOps/s 47.4355 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[False-False-False-True-True] 0.1395ms 57.5525μs 17.3754 KOps/s 17.1577 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[False-False-False-True-False] 75.9740μs 38.5017μs 25.9729 KOps/s 25.9936 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[False-False-False-False-True] 0.1166ms 36.5006μs 27.3968 KOps/s 27.6726 KOps/s $\color{#d91a1a}-1.00\%$
test_step_mdp_speed[False-False-False-False-False] 0.1028ms 23.4969μs 42.5589 KOps/s 42.6434 KOps/s $\color{#d91a1a}-0.20\%$
test_values[generalized_advantage_estimate-True-True] 10.2490ms 10.1245ms 98.7702 Ops/s 101.8649 Ops/s $\color{#d91a1a}-3.04\%$
test_values[vec_generalized_advantage_estimate-True-True] 20.6842ms 17.8371ms 56.0629 Ops/s 86.7633 Ops/s $\textbf{\color{#d91a1a}-35.38\%}$
test_values[td0_return_estimate-False-False] 0.2223ms 0.1326ms 7.5435 KOps/s 7.8263 KOps/s $\color{#d91a1a}-3.61\%$
test_values[td1_return_estimate-False-False] 27.9058ms 27.4433ms 36.4388 Ops/s 37.2835 Ops/s $\color{#d91a1a}-2.27\%$
test_values[vec_td1_return_estimate-False-False] 18.0711ms 17.8411ms 56.0504 Ops/s 87.8973 Ops/s $\textbf{\color{#d91a1a}-36.23\%}$
test_values[td_lambda_return_estimate-True-False] 41.7032ms 41.3577ms 24.1793 Ops/s 24.8677 Ops/s $\color{#d91a1a}-2.77\%$
test_values[vec_td_lambda_return_estimate-True-False] 17.9993ms 17.7403ms 56.3688 Ops/s 88.4157 Ops/s $\textbf{\color{#d91a1a}-36.25\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.6675ms 8.5849ms 116.4832 Ops/s 118.1849 Ops/s $\color{#d91a1a}-1.44\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7354ms 1.5227ms 656.7447 Ops/s 676.1419 Ops/s $\color{#d91a1a}-2.87\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4702ms 0.4205ms 2.3779 KOps/s 2.4206 KOps/s $\color{#d91a1a}-1.76\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.6857ms 33.9529ms 29.4526 Ops/s 29.7912 Ops/s $\color{#d91a1a}-1.14\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9020ms 1.7416ms 574.1839 Ops/s 567.9031 Ops/s $\color{#35bf28}+1.11\%$
test_dqn_speed[False-None] 1.5575ms 1.4609ms 684.5053 Ops/s 699.0428 Ops/s $\color{#d91a1a}-2.08\%$
test_dqn_speed[False-backward] 2.0777ms 1.9687ms 507.9466 Ops/s 516.1502 Ops/s $\color{#d91a1a}-1.59\%$
test_dqn_speed[True-None] 0.6555ms 0.5253ms 1.9036 KOps/s 1.9207 KOps/s $\color{#d91a1a}-0.89\%$
test_dqn_speed[True-backward] 1.0699ms 0.9778ms 1.0227 KOps/s 1.0107 KOps/s $\color{#35bf28}+1.19\%$
test_dqn_speed[reduce-overhead-None] 0.6360ms 0.5180ms 1.9304 KOps/s 2.0099 KOps/s $\color{#d91a1a}-3.95\%$
test_dqn_speed[reduce-overhead-backward] 1.0170ms 0.9549ms 1.0472 KOps/s 939.7551 Ops/s $\textbf{\color{#35bf28}+11.44\%}$
test_ddpg_speed[False-None] 3.2927ms 2.9218ms 342.2548 Ops/s 345.3020 Ops/s $\color{#d91a1a}-0.88\%$
test_ddpg_speed[False-backward] 6.0885ms 4.2465ms 235.4868 Ops/s 241.9573 Ops/s $\color{#d91a1a}-2.67\%$
test_ddpg_speed[True-None] 1.6931ms 1.3883ms 720.2922 Ops/s 719.0759 Ops/s $\color{#35bf28}+0.17\%$
test_ddpg_speed[True-backward] 2.4124ms 2.3619ms 423.3915 Ops/s 425.0106 Ops/s $\color{#d91a1a}-0.38\%$
test_ddpg_speed[reduce-overhead-None] 1.5214ms 1.3821ms 723.5330 Ops/s 738.1863 Ops/s $\color{#d91a1a}-1.99\%$
test_ddpg_speed[reduce-overhead-backward] 2.5181ms 2.4152ms 414.0433 Ops/s 409.1253 Ops/s $\color{#35bf28}+1.20\%$
test_sac_speed[False-None] 8.4950ms 7.8746ms 126.9902 Ops/s 128.1505 Ops/s $\color{#d91a1a}-0.91\%$
test_sac_speed[False-backward] 11.8548ms 11.2036ms 89.2569 Ops/s 90.2368 Ops/s $\color{#d91a1a}-1.09\%$
test_sac_speed[True-None] 2.4226ms 2.1003ms 476.1193 Ops/s 470.3452 Ops/s $\color{#35bf28}+1.23\%$
test_sac_speed[True-backward] 4.1236ms 3.9759ms 251.5139 Ops/s 240.7976 Ops/s $\color{#35bf28}+4.45\%$
test_sac_speed[reduce-overhead-None] 2.2728ms 2.1115ms 473.5914 Ops/s 472.4937 Ops/s $\color{#35bf28}+0.23\%$
test_sac_speed[reduce-overhead-backward] 4.5542ms 4.0119ms 249.2602 Ops/s 240.0251 Ops/s $\color{#35bf28}+3.85\%$
test_redq_speed[False-None] 16.7498ms 12.0054ms 83.2956 Ops/s 94.4427 Ops/s $\textbf{\color{#d91a1a}-11.80\%}$
test_redq_speed[False-backward] 25.2704ms 19.4511ms 51.4110 Ops/s 55.4636 Ops/s $\textbf{\color{#d91a1a}-7.31\%}$
test_redq_speed[True-None] 4.6324ms 4.3199ms 231.4865 Ops/s 226.8435 Ops/s $\color{#35bf28}+2.05\%$
test_redq_speed[True-backward] 10.0611ms 9.7386ms 102.6843 Ops/s 103.7008 Ops/s $\color{#d91a1a}-0.98\%$
test_redq_speed[reduce-overhead-None] 4.4789ms 4.2853ms 233.3569 Ops/s 226.1644 Ops/s $\color{#35bf28}+3.18\%$
test_redq_speed[reduce-overhead-backward] 10.1063ms 9.8242ms 101.7898 Ops/s 97.0287 Ops/s $\color{#35bf28}+4.91\%$
test_redq_deprec_speed[False-None] 11.4048ms 11.0070ms 90.8512 Ops/s 88.5244 Ops/s $\color{#35bf28}+2.63\%$
test_redq_deprec_speed[False-backward] 16.5212ms 15.9848ms 62.5594 Ops/s 61.8482 Ops/s $\color{#35bf28}+1.15\%$
test_redq_deprec_speed[True-None] 3.7741ms 3.6158ms 276.5610 Ops/s 272.7232 Ops/s $\color{#35bf28}+1.41\%$
test_redq_deprec_speed[True-backward] 7.9338ms 7.7380ms 129.2327 Ops/s 135.9573 Ops/s $\color{#d91a1a}-4.95\%$
test_redq_deprec_speed[reduce-overhead-None] 3.7346ms 3.5166ms 284.3663 Ops/s 289.2591 Ops/s $\color{#d91a1a}-1.69\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.7939ms 7.5553ms 132.3578 Ops/s 123.8662 Ops/s $\textbf{\color{#35bf28}+6.86\%}$
test_td3_speed[False-None] 8.0370ms 7.9429ms 125.8985 Ops/s 119.6256 Ops/s $\textbf{\color{#35bf28}+5.24\%}$
test_td3_speed[False-backward] 11.3376ms 10.7705ms 92.8463 Ops/s 91.6497 Ops/s $\color{#35bf28}+1.31\%$
test_td3_speed[True-None] 1.8907ms 1.7961ms 556.7602 Ops/s 558.3466 Ops/s $\color{#d91a1a}-0.28\%$
test_td3_speed[True-backward] 3.6480ms 3.5456ms 282.0434 Ops/s 275.8148 Ops/s $\color{#35bf28}+2.26\%$
test_td3_speed[reduce-overhead-None] 1.8154ms 1.7645ms 566.7274 Ops/s 567.5743 Ops/s $\color{#d91a1a}-0.15\%$
test_td3_speed[reduce-overhead-backward] 3.6779ms 3.5677ms 280.2952 Ops/s 231.3349 Ops/s $\textbf{\color{#35bf28}+21.16\%}$
test_cql_speed[False-None] 29.3949ms 26.1152ms 38.2919 Ops/s 37.9438 Ops/s $\color{#35bf28}+0.92\%$
test_cql_speed[False-backward] 39.7979ms 35.7776ms 27.9505 Ops/s 28.4912 Ops/s $\color{#d91a1a}-1.90\%$
test_cql_speed[True-None] 13.1597ms 12.3867ms 80.7318 Ops/s 79.0399 Ops/s $\color{#35bf28}+2.14\%$
test_cql_speed[True-backward] 18.8544ms 18.5099ms 54.0250 Ops/s 57.0409 Ops/s $\textbf{\color{#d91a1a}-5.29\%}$
test_cql_speed[reduce-overhead-None] 13.3878ms 12.4376ms 80.4016 Ops/s 81.0868 Ops/s $\color{#d91a1a}-0.85\%$
test_cql_speed[reduce-overhead-backward] 18.9790ms 18.4116ms 54.3137 Ops/s 56.4802 Ops/s $\color{#d91a1a}-3.84\%$
test_a2c_speed[False-None] 5.5948ms 5.3571ms 186.6668 Ops/s 181.3721 Ops/s $\color{#35bf28}+2.92\%$
test_a2c_speed[False-backward] 11.9534ms 11.6176ms 86.0762 Ops/s 83.7133 Ops/s $\color{#35bf28}+2.82\%$
test_a2c_speed[True-None] 3.9487ms 3.6861ms 271.2891 Ops/s 273.3408 Ops/s $\color{#d91a1a}-0.75\%$
test_a2c_speed[True-backward] 8.6983ms 8.4215ms 118.7444 Ops/s 110.8743 Ops/s $\textbf{\color{#35bf28}+7.10\%}$
test_a2c_speed[reduce-overhead-None] 3.8452ms 3.6899ms 271.0123 Ops/s 271.2279 Ops/s $\color{#d91a1a}-0.08\%$
test_a2c_speed[reduce-overhead-backward] 9.2915ms 8.8598ms 112.8695 Ops/s 113.4961 Ops/s $\color{#d91a1a}-0.55\%$
test_ppo_speed[False-None] 6.1226ms 5.8852ms 169.9190 Ops/s 172.0307 Ops/s $\color{#d91a1a}-1.23\%$
test_ppo_speed[False-backward] 12.8390ms 12.4492ms 80.3266 Ops/s 79.6677 Ops/s $\color{#35bf28}+0.83\%$
test_ppo_speed[True-None] 3.8136ms 3.6306ms 275.4370 Ops/s 274.5355 Ops/s $\color{#35bf28}+0.33\%$
test_ppo_speed[True-backward] 8.8561ms 8.5946ms 116.3528 Ops/s 114.6360 Ops/s $\color{#35bf28}+1.50\%$
test_ppo_speed[reduce-overhead-None] 3.7930ms 3.6069ms 277.2437 Ops/s 276.0320 Ops/s $\color{#35bf28}+0.44\%$
test_ppo_speed[reduce-overhead-backward] 8.9457ms 8.6169ms 116.0512 Ops/s 111.2619 Ops/s $\color{#35bf28}+4.30\%$
test_reinforce_speed[False-None] 4.9346ms 4.6683ms 214.2113 Ops/s 216.2783 Ops/s $\color{#d91a1a}-0.96\%$
test_reinforce_speed[False-backward] 7.6605ms 7.4929ms 133.4594 Ops/s 135.3106 Ops/s $\color{#d91a1a}-1.37\%$
test_reinforce_speed[True-None] 3.2178ms 2.8887ms 346.1729 Ops/s 352.0547 Ops/s $\color{#d91a1a}-1.67\%$
test_reinforce_speed[True-backward] 7.9182ms 7.6453ms 130.7991 Ops/s 128.3034 Ops/s $\color{#35bf28}+1.95\%$
test_reinforce_speed[reduce-overhead-None] 3.4016ms 2.8498ms 350.9032 Ops/s 339.3677 Ops/s $\color{#35bf28}+3.40\%$
test_reinforce_speed[reduce-overhead-backward] 8.1963ms 7.9271ms 126.1490 Ops/s 125.7445 Ops/s $\color{#35bf28}+0.32\%$
test_iql_speed[False-None] 25.8402ms 20.6302ms 48.4727 Ops/s 49.2034 Ops/s $\color{#d91a1a}-1.49\%$
test_iql_speed[False-backward] 31.1945ms 30.2804ms 33.0247 Ops/s 32.4188 Ops/s $\color{#35bf28}+1.87\%$
test_iql_speed[True-None] 8.8052ms 8.4178ms 118.7961 Ops/s 118.1617 Ops/s $\color{#35bf28}+0.54\%$
test_iql_speed[True-backward] 17.0614ms 16.6530ms 60.0492 Ops/s 59.6657 Ops/s $\color{#35bf28}+0.64\%$
test_iql_speed[reduce-overhead-None] 10.9329ms 8.5551ms 116.8890 Ops/s 116.2048 Ops/s $\color{#35bf28}+0.59\%$
test_iql_speed[reduce-overhead-backward] 17.3970ms 17.0832ms 58.5371 Ops/s 58.3457 Ops/s $\color{#35bf28}+0.33\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5194ms 6.0681ms 164.7975 Ops/s 163.0996 Ops/s $\color{#35bf28}+1.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5371ms 0.2753ms 3.6329 KOps/s 3.4102 KOps/s $\textbf{\color{#35bf28}+6.53\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5058ms 0.2568ms 3.8942 KOps/s 3.5292 KOps/s $\textbf{\color{#35bf28}+10.34\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9817ms 5.7550ms 173.7611 Ops/s 175.5386 Ops/s $\color{#d91a1a}-1.01\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8403ms 0.2907ms 3.4398 KOps/s 2.9407 KOps/s $\textbf{\color{#35bf28}+16.97\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5780ms 0.2870ms 3.4846 KOps/s 3.0868 KOps/s $\textbf{\color{#35bf28}+12.89\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5887ms 1.3596ms 735.4857 Ops/s 811.1990 Ops/s $\textbf{\color{#d91a1a}-9.33\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5689ms 1.2492ms 800.5369 Ops/s 856.9760 Ops/s $\textbf{\color{#d91a1a}-6.59\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.7827ms 6.0481ms 165.3417 Ops/s 168.0268 Ops/s $\color{#d91a1a}-1.60\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2489ms 0.4512ms 2.2165 KOps/s 2.0616 KOps/s $\textbf{\color{#35bf28}+7.52\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6644ms 0.4355ms 2.2965 KOps/s 2.1574 KOps/s $\textbf{\color{#35bf28}+6.45\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9630ms 5.7735ms 173.2060 Ops/s 171.1369 Ops/s $\color{#35bf28}+1.21\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.1169ms 0.3229ms 3.0972 KOps/s 829.9205 Ops/s $\textbf{\color{#35bf28}+273.20\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5061ms 0.2867ms 3.4882 KOps/s 3.0175 KOps/s $\textbf{\color{#35bf28}+15.60\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9908ms 5.7212ms 174.7892 Ops/s 174.3405 Ops/s $\color{#35bf28}+0.26\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8524ms 0.2893ms 3.4565 KOps/s 2.9484 KOps/s $\textbf{\color{#35bf28}+17.23\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4472ms 0.2566ms 3.8971 KOps/s 3.0613 KOps/s $\textbf{\color{#35bf28}+27.30\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.9949ms 5.9120ms 169.1474 Ops/s 168.9618 Ops/s $\color{#35bf28}+0.11\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1342ms 0.4274ms 2.3396 KOps/s 2.0602 KOps/s $\textbf{\color{#35bf28}+13.56\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6876ms 0.4497ms 2.2238 KOps/s 2.1251 KOps/s $\color{#35bf28}+4.64\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.6761ms 5.1491ms 194.2082 Ops/s 193.7195 Ops/s $\color{#35bf28}+0.25\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.2644ms 2.3034ms 434.1476 Ops/s 439.4814 Ops/s $\color{#d91a1a}-1.21\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.2468ms 1.2008ms 832.7857 Ops/s 846.6129 Ops/s $\color{#d91a1a}-1.63\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5060s 15.0788ms 66.3181 Ops/s 55.6439 Ops/s $\textbf{\color{#35bf28}+19.18\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.7486ms 1.9294ms 518.3039 Ops/s 482.9148 Ops/s $\textbf{\color{#35bf28}+7.33\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.6309ms 1.1722ms 853.0927 Ops/s 909.3062 Ops/s $\textbf{\color{#d91a1a}-6.18\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.0063ms 5.2783ms 189.4554 Ops/s 184.1003 Ops/s $\color{#35bf28}+2.91\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.2696ms 2.1912ms 456.3757 Ops/s 461.1175 Ops/s $\color{#d91a1a}-1.03\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.1738ms 1.3165ms 759.5693 Ops/s 743.2301 Ops/s $\color{#35bf28}+2.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.7102ms 32.6233ms 30.6530 Ops/s 30.3447 Ops/s $\color{#35bf28}+1.02\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.6986ms 17.6631ms 56.6151 Ops/s 57.0024 Ops/s $\color{#d91a1a}-0.68\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 35.6771ms 33.8645ms 29.5294 Ops/s 29.3792 Ops/s $\color{#35bf28}+0.51\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.8024ms 18.3609ms 54.4636 Ops/s 56.3503 Ops/s $\color{#d91a1a}-3.35\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 36.9972ms 35.4354ms 28.2204 Ops/s 28.0113 Ops/s $\color{#35bf28}+0.75\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.1747ms 19.1885ms 52.1146 Ops/s 52.2510 Ops/s $\color{#d91a1a}-0.26\%$

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 25, 2025
ghstack-source-id: 21eb8cc
Pull-Request: #3223
vmoens added a commit that referenced this pull request Oct 25, 2025
ghstack-source-id: 21eb8cc
Pull-Request: #3223
@vmoens vmoens merged commit 09f9d27 into gh/vmoens/170/base Oct 25, 2025
69 of 89 checks passed
@vmoens vmoens deleted the gh/vmoens/170/head branch October 25, 2025 16:23
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.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant