Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Dec 12, 2025

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 12, 2025
ghstack-source-id: ee09d7f
Pull-Request: #3251
@pytorch-bot
Copy link

pytorch-bot bot commented Dec 12, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3251

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

❌ 7 New Failures, 8 Unrelated Failures

As of commit bcbfbed with merge base 2c42fe2 (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 Dec 12, 2025
@vmoens vmoens added the ciflow/binaries/all Build all binaries label Dec 12, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 12, 2025
ghstack-source-id: 3ec1686
Pull-Request: #3251
@vmoens vmoens mentioned this pull request Dec 12, 2025
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Dec 12, 2025

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

Total Benchmarks: 164. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}21$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 87.3266μs 85.5694μs 11.6864 KOps/s 11.8854 KOps/s $\color{#d91a1a}-1.67\%$
test_tensor_to_bytestream_speed[torch.save] 0.1445ms 0.1432ms 6.9823 KOps/s 7.1375 KOps/s $\color{#d91a1a}-2.17\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1179s 0.1171s 8.5400 Ops/s 8.5564 Ops/s $\color{#d91a1a}-0.19\%$
test_tensor_to_bytestream_speed[numpy] 2.7224μs 2.7134μs 368.5472 KOps/s 367.0308 KOps/s $\color{#35bf28}+0.41\%$
test_tensor_to_bytestream_speed[safetensors] 38.0838μs 37.8608μs 26.4125 KOps/s 26.0904 KOps/s $\color{#35bf28}+1.23\%$
test_simple 0.5782s 0.5705s 1.7528 Ops/s 1.7357 Ops/s $\color{#35bf28}+0.98\%$
test_transformed 1.1210s 1.1202s 0.8927 Ops/s 0.8702 Ops/s $\color{#35bf28}+2.59\%$
test_serial 1.6924s 1.6696s 0.5989 Ops/s 0.5908 Ops/s $\color{#35bf28}+1.38\%$
test_parallel 1.2858s 1.2387s 0.8073 Ops/s 0.8873 Ops/s $\textbf{\color{#d91a1a}-9.02\%}$
test_step_mdp_speed[True-True-True-True-True] 0.1733ms 46.0684μs 21.7068 KOps/s 21.5381 KOps/s $\color{#35bf28}+0.78\%$
test_step_mdp_speed[True-True-True-True-False] 57.4930μs 25.7230μs 38.8757 KOps/s 38.1743 KOps/s $\color{#35bf28}+1.84\%$
test_step_mdp_speed[True-True-True-False-True] 57.9430μs 25.5888μs 39.0796 KOps/s 38.4731 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[True-True-True-False-False] 45.7020μs 14.1350μs 70.7465 KOps/s 69.4781 KOps/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[True-True-False-True-True] 89.8850μs 48.7104μs 20.5295 KOps/s 20.1774 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-True-False-True-False] 61.2830μs 28.7288μs 34.8082 KOps/s 34.2298 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[True-True-False-False-True] 67.0040μs 28.1810μs 35.4849 KOps/s 34.3202 KOps/s $\color{#35bf28}+3.39\%$
test_step_mdp_speed[True-True-False-False-False] 68.0030μs 16.8571μs 59.3222 KOps/s 57.8165 KOps/s $\color{#35bf28}+2.60\%$
test_step_mdp_speed[True-False-True-True-True] 95.9860μs 51.3738μs 19.4652 KOps/s 19.0913 KOps/s $\color{#35bf28}+1.96\%$
test_step_mdp_speed[True-False-True-True-False] 73.8240μs 31.1549μs 32.0977 KOps/s 30.8316 KOps/s $\color{#35bf28}+4.11\%$
test_step_mdp_speed[True-False-True-False-True] 80.7450μs 28.3108μs 35.3222 KOps/s 34.7193 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-False-True-False-False] 52.2230μs 16.8463μs 59.3602 KOps/s 58.0797 KOps/s $\color{#35bf28}+2.20\%$
test_step_mdp_speed[True-False-False-True-True] 91.2450μs 53.4182μs 18.7202 KOps/s 18.2278 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[True-False-False-True-False] 68.4840μs 33.5467μs 29.8092 KOps/s 28.7677 KOps/s $\color{#35bf28}+3.62\%$
test_step_mdp_speed[True-False-False-False-True] 84.2150μs 31.0219μs 32.2353 KOps/s 31.6298 KOps/s $\color{#35bf28}+1.91\%$
test_step_mdp_speed[True-False-False-False-False] 59.4740μs 19.3348μs 51.7201 KOps/s 50.0633 KOps/s $\color{#35bf28}+3.31\%$
test_step_mdp_speed[False-True-True-True-True] 97.4550μs 51.6964μs 19.3437 KOps/s 19.1809 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[False-True-True-True-False] 72.2640μs 30.8711μs 32.3928 KOps/s 30.5064 KOps/s $\textbf{\color{#35bf28}+6.18\%}$
test_step_mdp_speed[False-True-True-False-True] 2.3445ms 32.3439μs 30.9178 KOps/s 30.2011 KOps/s $\color{#35bf28}+2.37\%$
test_step_mdp_speed[False-True-True-False-False] 75.3540μs 17.7785μs 56.2477 KOps/s 52.5964 KOps/s $\textbf{\color{#35bf28}+6.94\%}$
test_step_mdp_speed[False-True-False-True-True] 0.1359ms 53.1873μs 18.8015 KOps/s 18.0606 KOps/s $\color{#35bf28}+4.10\%$
test_step_mdp_speed[False-True-False-True-False] 63.8440μs 33.0038μs 30.2996 KOps/s 28.1512 KOps/s $\textbf{\color{#35bf28}+7.63\%}$
test_step_mdp_speed[False-True-False-False-True] 69.3440μs 34.3429μs 29.1181 KOps/s 28.0311 KOps/s $\color{#35bf28}+3.88\%$
test_step_mdp_speed[False-True-False-False-False] 48.0520μs 21.3773μs 46.7785 KOps/s 44.7777 KOps/s $\color{#35bf28}+4.47\%$
test_step_mdp_speed[False-False-True-True-True] 96.8960μs 55.4561μs 18.0323 KOps/s 17.0784 KOps/s $\textbf{\color{#35bf28}+5.59\%}$
test_step_mdp_speed[False-False-True-True-False] 64.6340μs 36.4669μs 27.4221 KOps/s 26.1542 KOps/s $\color{#35bf28}+4.85\%$
test_step_mdp_speed[False-False-True-False-True] 59.4240μs 34.2321μs 29.2123 KOps/s 27.8983 KOps/s $\color{#35bf28}+4.71\%$
test_step_mdp_speed[False-False-True-False-False] 50.0130μs 20.8763μs 47.9012 KOps/s 44.2539 KOps/s $\textbf{\color{#35bf28}+8.24\%}$
test_step_mdp_speed[False-False-False-True-True] 92.8560μs 58.0535μs 17.2255 KOps/s 16.5924 KOps/s $\color{#35bf28}+3.82\%$
test_step_mdp_speed[False-False-False-True-False] 70.4450μs 37.7526μs 26.4882 KOps/s 24.3276 KOps/s $\textbf{\color{#35bf28}+8.88\%}$
test_step_mdp_speed[False-False-False-False-True] 0.1031ms 36.6180μs 27.3090 KOps/s 26.3246 KOps/s $\color{#35bf28}+3.74\%$
test_step_mdp_speed[False-False-False-False-False] 65.5740μs 23.3463μs 42.8333 KOps/s 40.6336 KOps/s $\textbf{\color{#35bf28}+5.41\%}$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8860s 0.7839s 1.2757 Ops/s 1.2731 Ops/s $\color{#35bf28}+0.21\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7425s 0.6458s 1.5485 Ops/s 1.5437 Ops/s $\color{#35bf28}+0.32\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7915s 1.7090s 0.5852 Ops/s 0.5844 Ops/s $\color{#35bf28}+0.12\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5600s 1.4763s 0.6774 Ops/s 0.6725 Ops/s $\color{#35bf28}+0.73\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0396s 1.9506s 0.5127 Ops/s 0.5114 Ops/s $\color{#35bf28}+0.25\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.8090s 1.7269s 0.5791 Ops/s 0.5762 Ops/s $\color{#35bf28}+0.49\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.7511s 4.6821s 0.2136 Ops/s 0.2157 Ops/s $\color{#d91a1a}-0.99\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.5488s 4.4554s 0.2244 Ops/s 0.2223 Ops/s $\color{#35bf28}+0.98\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0620s 1.9961s 0.5010 Ops/s 0.5040 Ops/s $\color{#d91a1a}-0.59\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7594s 1.6801s 0.5952 Ops/s 0.5890 Ops/s $\color{#35bf28}+1.05\%$
test_values[generalized_advantage_estimate-True-True] 10.0075ms 9.8595ms 101.4248 Ops/s 97.0607 Ops/s $\color{#35bf28}+4.50\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.9273ms 17.6939ms 56.5168 Ops/s 88.2238 Ops/s $\textbf{\color{#d91a1a}-35.94\%}$
test_values[td0_return_estimate-False-False] 0.2202ms 0.1263ms 7.9200 KOps/s 7.7048 KOps/s $\color{#35bf28}+2.79\%$
test_values[td1_return_estimate-False-False] 26.9820ms 26.2693ms 38.0673 Ops/s 37.2159 Ops/s $\color{#35bf28}+2.29\%$
test_values[vec_td1_return_estimate-False-False] 18.4345ms 17.7765ms 56.2540 Ops/s 88.1357 Ops/s $\textbf{\color{#d91a1a}-36.17\%}$
test_values[td_lambda_return_estimate-True-False] 39.6847ms 38.7458ms 25.8092 Ops/s 25.2444 Ops/s $\color{#35bf28}+2.24\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.8059ms 17.9927ms 55.5780 Ops/s 87.1684 Ops/s $\textbf{\color{#d91a1a}-36.24\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.9790ms 8.8910ms 112.4730 Ops/s 108.8324 Ops/s $\color{#35bf28}+3.35\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.6859ms 1.4986ms 667.2917 Ops/s 665.6987 Ops/s $\color{#35bf28}+0.24\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4688ms 0.4121ms 2.4264 KOps/s 2.3799 KOps/s $\color{#35bf28}+1.95\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.9810ms 35.1412ms 28.4566 Ops/s 40.7285 Ops/s $\textbf{\color{#d91a1a}-30.13\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.0142ms 1.7128ms 583.8359 Ops/s 582.0402 Ops/s $\color{#35bf28}+0.31\%$
test_dqn_speed[False-None] 1.6824ms 1.3974ms 715.6087 Ops/s 694.0084 Ops/s $\color{#35bf28}+3.11\%$
test_dqn_speed[False-backward] 2.0130ms 1.9017ms 525.8524 Ops/s 514.5338 Ops/s $\color{#35bf28}+2.20\%$
test_dqn_speed[True-None] 0.5918ms 0.5275ms 1.8958 KOps/s 1.8223 KOps/s $\color{#35bf28}+4.03\%$
test_dqn_speed[True-backward] 1.0567ms 1.0018ms 998.2414 Ops/s 896.2921 Ops/s $\textbf{\color{#35bf28}+11.37\%}$
test_dqn_speed[reduce-overhead-None] 0.6184ms 0.5342ms 1.8720 KOps/s 1.8083 KOps/s $\color{#35bf28}+3.52\%$
test_dqn_speed[reduce-overhead-backward] 1.0225ms 0.9893ms 1.0108 KOps/s 871.0678 Ops/s $\textbf{\color{#35bf28}+16.05\%}$
test_ddpg_speed[False-None] 3.5290ms 2.8470ms 351.2439 Ops/s 340.0757 Ops/s $\color{#35bf28}+3.28\%$
test_ddpg_speed[False-backward] 4.1684ms 4.0615ms 246.2117 Ops/s 246.7581 Ops/s $\color{#d91a1a}-0.22\%$
test_ddpg_speed[True-None] 1.7521ms 1.4175ms 705.4528 Ops/s 703.2979 Ops/s $\color{#35bf28}+0.31\%$
test_ddpg_speed[True-backward] 2.4711ms 2.4019ms 416.3309 Ops/s 367.8994 Ops/s $\textbf{\color{#35bf28}+13.16\%}$
test_ddpg_speed[reduce-overhead-None] 1.4822ms 1.3985ms 715.0637 Ops/s 703.2209 Ops/s $\color{#35bf28}+1.68\%$
test_ddpg_speed[reduce-overhead-backward] 2.4515ms 2.4047ms 415.8549 Ops/s 406.9436 Ops/s $\color{#35bf28}+2.19\%$
test_sac_speed[False-None] 9.3557ms 7.9420ms 125.9133 Ops/s 126.5342 Ops/s $\color{#d91a1a}-0.49\%$
test_sac_speed[False-backward] 11.3802ms 11.1406ms 89.7617 Ops/s 87.6862 Ops/s $\color{#35bf28}+2.37\%$
test_sac_speed[True-None] 2.3294ms 2.1773ms 459.2929 Ops/s 460.5473 Ops/s $\color{#d91a1a}-0.27\%$
test_sac_speed[True-backward] 4.2245ms 4.0961ms 244.1336 Ops/s 234.3989 Ops/s $\color{#35bf28}+4.15\%$
test_sac_speed[reduce-overhead-None] 2.4454ms 2.2086ms 452.7697 Ops/s 446.4715 Ops/s $\color{#35bf28}+1.41\%$
test_sac_speed[reduce-overhead-backward] 4.3360ms 4.1373ms 241.7052 Ops/s 236.4637 Ops/s $\color{#35bf28}+2.22\%$
test_redq_speed[False-None] 16.1745ms 10.7434ms 93.0808 Ops/s 95.2967 Ops/s $\color{#d91a1a}-2.33\%$
test_redq_speed[False-backward] 20.2995ms 18.5829ms 53.8128 Ops/s 55.5073 Ops/s $\color{#d91a1a}-3.05\%$
test_redq_speed[True-None] 4.8054ms 4.5560ms 219.4904 Ops/s 210.2922 Ops/s $\color{#35bf28}+4.37\%$
test_redq_speed[True-backward] 10.6142ms 10.1601ms 98.4244 Ops/s 99.3416 Ops/s $\color{#d91a1a}-0.92\%$
test_redq_speed[reduce-overhead-None] 4.9069ms 4.6551ms 214.8201 Ops/s 214.3747 Ops/s $\color{#35bf28}+0.21\%$
test_redq_speed[reduce-overhead-backward] 10.8941ms 10.4142ms 96.0231 Ops/s 98.5262 Ops/s $\color{#d91a1a}-2.54\%$
test_redq_deprec_speed[False-None] 13.8085ms 11.1982ms 89.2999 Ops/s 87.8970 Ops/s $\color{#35bf28}+1.60\%$
test_redq_deprec_speed[False-backward] 16.8757ms 16.2279ms 61.6223 Ops/s 60.8265 Ops/s $\color{#35bf28}+1.31\%$
test_redq_deprec_speed[True-None] 4.2248ms 3.7632ms 265.7292 Ops/s 260.1105 Ops/s $\color{#35bf28}+2.16\%$
test_redq_deprec_speed[True-backward] 8.0743ms 7.8447ms 127.4748 Ops/s 123.6338 Ops/s $\color{#35bf28}+3.11\%$
test_redq_deprec_speed[reduce-overhead-None] 3.9527ms 3.7309ms 268.0349 Ops/s 260.3363 Ops/s $\color{#35bf28}+2.96\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.4316ms 7.8891ms 126.7565 Ops/s 121.7122 Ops/s $\color{#35bf28}+4.14\%$
test_td3_speed[False-None] 8.2209ms 8.0041ms 124.9357 Ops/s 124.4143 Ops/s $\color{#35bf28}+0.42\%$
test_td3_speed[False-backward] 11.3702ms 10.8232ms 92.3941 Ops/s 92.1687 Ops/s $\color{#35bf28}+0.24\%$
test_td3_speed[True-None] 1.9213ms 1.8851ms 530.4889 Ops/s 518.9148 Ops/s $\color{#35bf28}+2.23\%$
test_td3_speed[True-backward] 3.8949ms 3.7762ms 264.8139 Ops/s 267.0585 Ops/s $\color{#d91a1a}-0.84\%$
test_td3_speed[reduce-overhead-None] 1.9116ms 1.8553ms 538.9920 Ops/s 542.4623 Ops/s $\color{#d91a1a}-0.64\%$
test_td3_speed[reduce-overhead-backward] 4.0926ms 3.7796ms 264.5763 Ops/s 247.1586 Ops/s $\textbf{\color{#35bf28}+7.05\%}$
test_cql_speed[False-None] 29.3044ms 26.4949ms 37.7432 Ops/s 37.7742 Ops/s $\color{#d91a1a}-0.08\%$
test_cql_speed[False-backward] 36.7901ms 35.8714ms 27.8773 Ops/s 27.9285 Ops/s $\color{#d91a1a}-0.18\%$
test_cql_speed[True-None] 13.0677ms 12.5726ms 79.5381 Ops/s 77.2918 Ops/s $\color{#35bf28}+2.91\%$
test_cql_speed[True-backward] 19.1035ms 18.7029ms 53.4677 Ops/s 55.2792 Ops/s $\color{#d91a1a}-3.28\%$
test_cql_speed[reduce-overhead-None] 13.2864ms 12.8063ms 78.0867 Ops/s 77.6824 Ops/s $\color{#35bf28}+0.52\%$
test_cql_speed[reduce-overhead-backward] 19.7932ms 18.8575ms 53.0294 Ops/s 54.9353 Ops/s $\color{#d91a1a}-3.47\%$
test_a2c_speed[False-None] 5.8233ms 5.4697ms 182.8261 Ops/s 180.3845 Ops/s $\color{#35bf28}+1.35\%$
test_a2c_speed[False-backward] 12.5549ms 11.9350ms 83.7875 Ops/s 83.0432 Ops/s $\color{#35bf28}+0.90\%$
test_a2c_speed[True-None] 4.0189ms 3.7601ms 265.9508 Ops/s 256.1928 Ops/s $\color{#35bf28}+3.81\%$
test_a2c_speed[True-backward] 9.1414ms 8.7996ms 113.6420 Ops/s 113.4232 Ops/s $\color{#35bf28}+0.19\%$
test_a2c_speed[reduce-overhead-None] 3.9177ms 3.7772ms 264.7455 Ops/s 263.9939 Ops/s $\color{#35bf28}+0.28\%$
test_a2c_speed[reduce-overhead-backward] 9.3790ms 8.9827ms 111.3254 Ops/s 107.7018 Ops/s $\color{#35bf28}+3.36\%$
test_ppo_speed[False-None] 6.2257ms 5.9645ms 167.6579 Ops/s 169.8826 Ops/s $\color{#d91a1a}-1.31\%$
test_ppo_speed[False-backward] 13.1026ms 12.8285ms 77.9517 Ops/s 81.1134 Ops/s $\color{#d91a1a}-3.90\%$
test_ppo_speed[True-None] 4.0956ms 3.6914ms 270.9004 Ops/s 260.1206 Ops/s $\color{#35bf28}+4.14\%$
test_ppo_speed[True-backward] 9.0992ms 8.6237ms 115.9592 Ops/s 116.2129 Ops/s $\color{#d91a1a}-0.22\%$
test_ppo_speed[reduce-overhead-None] 3.9869ms 3.6601ms 273.2180 Ops/s 272.8187 Ops/s $\color{#35bf28}+0.15\%$
test_ppo_speed[reduce-overhead-backward] 9.1084ms 8.8841ms 112.5600 Ops/s 107.2910 Ops/s $\color{#35bf28}+4.91\%$
test_reinforce_speed[False-None] 7.2635ms 4.6170ms 216.5887 Ops/s 215.4387 Ops/s $\color{#35bf28}+0.53\%$
test_reinforce_speed[False-backward] 7.8670ms 7.4683ms 133.8993 Ops/s 132.3541 Ops/s $\color{#35bf28}+1.17\%$
test_reinforce_speed[True-None] 3.2830ms 2.9201ms 342.4524 Ops/s 312.8144 Ops/s $\textbf{\color{#35bf28}+9.47\%}$
test_reinforce_speed[True-backward] 8.4358ms 7.8809ms 126.8899 Ops/s 121.2003 Ops/s $\color{#35bf28}+4.69\%$
test_reinforce_speed[reduce-overhead-None] 3.4946ms 2.9005ms 344.7717 Ops/s 341.0159 Ops/s $\color{#35bf28}+1.10\%$
test_reinforce_speed[reduce-overhead-backward] 9.9113ms 8.4684ms 118.0864 Ops/s 117.8278 Ops/s $\color{#35bf28}+0.22\%$
test_iql_speed[False-None] 25.5383ms 20.8143ms 48.0439 Ops/s 48.7220 Ops/s $\color{#d91a1a}-1.39\%$
test_iql_speed[False-backward] 32.1046ms 31.1938ms 32.0576 Ops/s 32.2038 Ops/s $\color{#d91a1a}-0.45\%$
test_iql_speed[True-None] 9.1448ms 8.8106ms 113.5002 Ops/s 110.8172 Ops/s $\color{#35bf28}+2.42\%$
test_iql_speed[True-backward] 17.7260ms 17.3118ms 57.7642 Ops/s 58.2082 Ops/s $\color{#d91a1a}-0.76\%$
test_iql_speed[reduce-overhead-None] 9.1884ms 8.9185ms 112.1260 Ops/s 111.6816 Ops/s $\color{#35bf28}+0.40\%$
test_iql_speed[reduce-overhead-backward] 18.3538ms 17.7929ms 56.2023 Ops/s 56.7603 Ops/s $\color{#d91a1a}-0.98\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2558ms 6.0962ms 164.0356 Ops/s 165.9665 Ops/s $\color{#d91a1a}-1.16\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9833ms 0.3667ms 2.7268 KOps/s 3.1986 KOps/s $\textbf{\color{#d91a1a}-14.75\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7117ms 0.3655ms 2.7361 KOps/s 3.7655 KOps/s $\textbf{\color{#d91a1a}-27.34\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0424ms 5.8115ms 172.0739 Ops/s 173.7512 Ops/s $\color{#d91a1a}-0.97\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.5344ms 0.3762ms 2.6582 KOps/s 3.5695 KOps/s $\textbf{\color{#d91a1a}-25.53\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6378ms 0.3584ms 2.7901 KOps/s 3.8377 KOps/s $\textbf{\color{#d91a1a}-27.30\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7807ms 1.4142ms 707.1106 Ops/s 754.1094 Ops/s $\textbf{\color{#d91a1a}-6.23\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6395ms 1.3316ms 750.9605 Ops/s 847.2080 Ops/s $\textbf{\color{#d91a1a}-11.36\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0057ms 5.8998ms 169.4959 Ops/s 167.5187 Ops/s $\color{#35bf28}+1.18\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1231ms 0.5281ms 1.8937 KOps/s 2.1110 KOps/s $\textbf{\color{#d91a1a}-10.30\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7548ms 0.5089ms 1.9652 KOps/s 2.2064 KOps/s $\textbf{\color{#d91a1a}-10.93\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8536ms 5.7762ms 173.1231 Ops/s 168.6373 Ops/s $\color{#35bf28}+2.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7761s 1.5667ms 638.2991 Ops/s 2.7363 KOps/s $\textbf{\color{#d91a1a}-76.67\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6461ms 0.3662ms 2.7306 KOps/s 3.7280 KOps/s $\textbf{\color{#d91a1a}-26.76\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0827ms 5.8528ms 170.8576 Ops/s 172.1704 Ops/s $\color{#d91a1a}-0.76\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1963ms 0.3768ms 2.6536 KOps/s 3.5201 KOps/s $\textbf{\color{#d91a1a}-24.61\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5577ms 0.3640ms 2.7469 KOps/s 3.7892 KOps/s $\textbf{\color{#d91a1a}-27.51\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1218ms 6.0024ms 166.6002 Ops/s 168.1516 Ops/s $\color{#d91a1a}-0.92\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9218ms 0.5300ms 1.8866 KOps/s 2.2261 KOps/s $\textbf{\color{#d91a1a}-15.25\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7267ms 0.5154ms 1.9401 KOps/s 2.3917 KOps/s $\textbf{\color{#d91a1a}-18.88\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.8595ms 5.0920ms 196.3863 Ops/s 195.8100 Ops/s $\color{#35bf28}+0.29\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.1790ms 2.4738ms 404.2346 Ops/s 429.7144 Ops/s $\textbf{\color{#d91a1a}-5.93\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.0686ms 1.1216ms 891.6064 Ops/s 798.0453 Ops/s $\textbf{\color{#35bf28}+11.72\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.6751s 18.5186ms 53.9998 Ops/s 48.8700 Ops/s $\textbf{\color{#35bf28}+10.50\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.0305ms 1.4698ms 680.3440 Ops/s 472.0678 Ops/s $\textbf{\color{#35bf28}+44.12\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 11.6536ms 1.3259ms 754.1952 Ops/s 818.6592 Ops/s $\textbf{\color{#d91a1a}-7.87\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.3017ms 5.3347ms 187.4511 Ops/s 187.5519 Ops/s $\color{#d91a1a}-0.05\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.0480ms 2.2293ms 448.5781 Ops/s 430.0811 Ops/s $\color{#35bf28}+4.30\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.5961ms 1.3952ms 716.7649 Ops/s 720.8939 Ops/s $\color{#d91a1a}-0.57\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.5748ms 33.3614ms 29.9748 Ops/s 29.6466 Ops/s $\color{#35bf28}+1.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.4979ms 17.6592ms 56.6276 Ops/s 57.4989 Ops/s $\color{#d91a1a}-1.52\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 37.5457ms 34.6325ms 28.8746 Ops/s 28.7994 Ops/s $\color{#35bf28}+0.26\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.9483ms 18.1113ms 55.2140 Ops/s 57.2505 Ops/s $\color{#d91a1a}-3.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 42.1392ms 36.4104ms 27.4647 Ops/s 26.9684 Ops/s $\color{#35bf28}+1.84\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.0251ms 19.2981ms 51.8185 Ops/s 52.8363 Ops/s $\color{#d91a1a}-1.93\%$

[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]
[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]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: c12afe8
Pull-Request: #3251
@vmoens vmoens merged commit bcbfbed into gh/vmoens/168/base Dec 18, 2025
243 of 258 checks passed
@vmoens vmoens deleted the gh/vmoens/168/head branch December 18, 2025 08:33
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ciflow/binaries/all Build all binaries 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.

2 participants