Skip to content

Conversation

[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/3185

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens added the enhancement New feature or request label Oct 19, 2025
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Oct 19, 2025

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

Total Benchmarks: 156. Improved: $\large\color{#35bf28}26$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.2112μs 81.5465μs 12.2629 KOps/s 11.5132 KOps/s $\textbf{\color{#35bf28}+6.51\%}$
test_tensor_to_bytestream_speed[torch.save] 0.1405ms 0.1401ms 7.1388 KOps/s 7.0809 KOps/s $\color{#35bf28}+0.82\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1164s 0.1155s 8.6571 Ops/s 8.7639 Ops/s $\color{#d91a1a}-1.22\%$
test_tensor_to_bytestream_speed[numpy] 2.8435μs 2.8274μs 353.6841 KOps/s 361.8523 KOps/s $\color{#d91a1a}-2.26\%$
test_tensor_to_bytestream_speed[safetensors] 42.4019μs 42.1683μs 23.7145 KOps/s 23.3873 KOps/s $\color{#35bf28}+1.40\%$
test_simple 0.6581s 0.5689s 1.7578 Ops/s 1.7552 Ops/s $\color{#35bf28}+0.15\%$
test_transformed 1.2155s 1.1257s 0.8883 Ops/s 0.8861 Ops/s $\color{#35bf28}+0.25\%$
test_serial 1.7554s 1.6700s 0.5988 Ops/s 0.5961 Ops/s $\color{#35bf28}+0.45\%$
test_parallel 1.1803s 1.0805s 0.9255 Ops/s 0.9104 Ops/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-True-True-True-True] 0.1642ms 44.3232μs 22.5616 KOps/s 22.2600 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[True-True-True-True-False] 55.4710μs 24.9751μs 40.0399 KOps/s 39.6582 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[True-True-True-False-True] 76.5120μs 24.9301μs 40.1121 KOps/s 38.8954 KOps/s $\color{#35bf28}+3.13\%$
test_step_mdp_speed[True-True-True-False-False] 51.8200μs 13.7614μs 72.6671 KOps/s 70.2785 KOps/s $\color{#35bf28}+3.40\%$
test_step_mdp_speed[True-True-False-True-True] 92.1710μs 47.2844μs 21.1486 KOps/s 21.0601 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[True-True-False-True-False] 57.9010μs 27.6389μs 36.1809 KOps/s 35.6553 KOps/s $\color{#35bf28}+1.47\%$
test_step_mdp_speed[True-True-False-False-True] 57.9210μs 28.1037μs 35.5825 KOps/s 35.8141 KOps/s $\color{#d91a1a}-0.65\%$
test_step_mdp_speed[True-True-False-False-False] 53.9210μs 16.7096μs 59.8460 KOps/s 60.1124 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[True-False-True-True-True] 86.2120μs 50.4571μs 19.8188 KOps/s 19.9441 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[True-False-True-True-False] 73.3210μs 30.3102μs 32.9922 KOps/s 32.5700 KOps/s $\color{#35bf28}+1.30\%$
test_step_mdp_speed[True-False-True-False-True] 64.4310μs 28.0652μs 35.6313 KOps/s 35.7415 KOps/s $\color{#d91a1a}-0.31\%$
test_step_mdp_speed[True-False-True-False-False] 44.9600μs 16.7449μs 59.7197 KOps/s 60.2009 KOps/s $\color{#d91a1a}-0.80\%$
test_step_mdp_speed[True-False-False-True-True] 86.9210μs 53.0781μs 18.8402 KOps/s 19.2484 KOps/s $\color{#d91a1a}-2.12\%$
test_step_mdp_speed[True-False-False-True-False] 67.5810μs 33.0594μs 30.2486 KOps/s 30.7725 KOps/s $\color{#d91a1a}-1.70\%$
test_step_mdp_speed[True-False-False-False-True] 86.2320μs 29.9045μs 33.4398 KOps/s 33.4146 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[True-False-False-False-False] 48.9310μs 19.1331μs 52.2654 KOps/s 51.9868 KOps/s $\color{#35bf28}+0.54\%$
test_step_mdp_speed[False-True-True-True-True] 91.7010μs 49.4648μs 20.2164 KOps/s 19.8771 KOps/s $\color{#35bf28}+1.71\%$
test_step_mdp_speed[False-True-True-True-False] 63.9710μs 30.3288μs 32.9720 KOps/s 32.7063 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[False-True-True-False-True] 2.4965ms 31.9555μs 31.2935 KOps/s 31.1825 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[False-True-True-False-False] 48.5100μs 18.1527μs 55.0883 KOps/s 53.3398 KOps/s $\color{#35bf28}+3.28\%$
test_step_mdp_speed[False-True-False-True-True] 97.2210μs 52.6023μs 19.0106 KOps/s 18.9860 KOps/s $\color{#35bf28}+0.13\%$
test_step_mdp_speed[False-True-False-True-False] 70.2310μs 33.0346μs 30.2713 KOps/s 30.7562 KOps/s $\color{#d91a1a}-1.58\%$
test_step_mdp_speed[False-True-False-False-True] 61.4810μs 34.2539μs 29.1938 KOps/s 29.5519 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[False-True-False-False-False] 61.0010μs 20.5283μs 48.7132 KOps/s 48.1001 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[False-False-True-True-True] 90.0710μs 55.2565μs 18.0974 KOps/s 18.2247 KOps/s $\color{#d91a1a}-0.70\%$
test_step_mdp_speed[False-False-True-True-False] 0.1032ms 35.7460μs 27.9752 KOps/s 28.3295 KOps/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[False-False-True-False-True] 64.8810μs 33.9857μs 29.4241 KOps/s 29.3500 KOps/s $\color{#35bf28}+0.25\%$
test_step_mdp_speed[False-False-True-False-False] 62.5010μs 20.4967μs 48.7884 KOps/s 49.0098 KOps/s $\color{#d91a1a}-0.45\%$
test_step_mdp_speed[False-False-False-True-True] 97.0020μs 57.1513μs 17.4974 KOps/s 17.8081 KOps/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[False-False-False-True-False] 0.1011ms 37.9378μs 26.3589 KOps/s 26.8744 KOps/s $\color{#d91a1a}-1.92\%$
test_step_mdp_speed[False-False-False-False-True] 78.5110μs 35.9580μs 27.8103 KOps/s 27.7440 KOps/s $\color{#35bf28}+0.24\%$
test_step_mdp_speed[False-False-False-False-False] 55.9510μs 23.1708μs 43.1577 KOps/s 43.3810 KOps/s $\color{#d91a1a}-0.51\%$
test_packing[True] 0.6511s 0.6444s 1.5518 Ops/s 1.5451 Ops/s $\color{#35bf28}+0.43\%$
test_packing[False] 0.7205s 0.7140s 1.4006 Ops/s 1.4003 Ops/s $\color{#35bf28}+0.03\%$
test_values[generalized_advantage_estimate-True-True] 9.8305ms 9.4760ms 105.5303 Ops/s 106.3115 Ops/s $\color{#d91a1a}-0.73\%$
test_values[vec_generalized_advantage_estimate-True-True] 12.4397ms 11.1221ms 89.9114 Ops/s 90.4955 Ops/s $\color{#d91a1a}-0.65\%$
test_values[td0_return_estimate-False-False] 0.2245ms 0.1291ms 7.7485 KOps/s 7.8777 KOps/s $\color{#d91a1a}-1.64\%$
test_values[td1_return_estimate-False-False] 26.2092ms 25.6683ms 38.9586 Ops/s 40.6708 Ops/s $\color{#d91a1a}-4.21\%$
test_values[vec_td1_return_estimate-False-False] 11.9250ms 11.0690ms 90.3426 Ops/s 90.0440 Ops/s $\color{#35bf28}+0.33\%$
test_values[td_lambda_return_estimate-True-False] 38.8877ms 38.1502ms 26.2122 Ops/s 26.9732 Ops/s $\color{#d91a1a}-2.82\%$
test_values[vec_td_lambda_return_estimate-True-False] 11.2360ms 11.0120ms 90.8104 Ops/s 90.1855 Ops/s $\color{#35bf28}+0.69\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.4294ms 8.2814ms 120.7530 Ops/s 123.2965 Ops/s $\color{#d91a1a}-2.06\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7557ms 1.5069ms 663.6009 Ops/s 656.5625 Ops/s $\color{#35bf28}+1.07\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4625ms 0.3946ms 2.5340 KOps/s 2.5745 KOps/s $\color{#d91a1a}-1.57\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 20.1792ms 19.6360ms 50.9268 Ops/s 51.9193 Ops/s $\color{#d91a1a}-1.91\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.0766ms 1.6960ms 589.6089 Ops/s 599.6581 Ops/s $\color{#d91a1a}-1.68\%$
test_dqn_speed[False-None] 1.7232ms 1.4060ms 711.2282 Ops/s 718.1145 Ops/s $\color{#d91a1a}-0.96\%$
test_dqn_speed[False-backward] 1.9353ms 1.8854ms 530.3977 Ops/s 541.8830 Ops/s $\color{#d91a1a}-2.12\%$
test_dqn_speed[True-None] 0.9182ms 0.5232ms 1.9113 KOps/s 1.8895 KOps/s $\color{#35bf28}+1.16\%$
test_dqn_speed[True-backward] 1.0886ms 0.9549ms 1.0473 KOps/s 904.7086 Ops/s $\textbf{\color{#35bf28}+15.76\%}$
test_dqn_speed[reduce-overhead-None] 0.8987ms 0.5089ms 1.9649 KOps/s 1.9152 KOps/s $\color{#35bf28}+2.59\%$
test_dqn_speed[reduce-overhead-backward] 0.9671ms 0.9415ms 1.0622 KOps/s 1.0882 KOps/s $\color{#d91a1a}-2.39\%$
test_ddpg_speed[False-None] 3.1324ms 2.8072ms 356.2216 Ops/s 353.3936 Ops/s $\color{#35bf28}+0.80\%$
test_ddpg_speed[False-backward] 4.1652ms 3.9911ms 250.5597 Ops/s 250.4698 Ops/s $\color{#35bf28}+0.04\%$
test_ddpg_speed[True-None] 1.6066ms 1.3983ms 715.1564 Ops/s 722.2421 Ops/s $\color{#d91a1a}-0.98\%$
test_ddpg_speed[True-backward] 2.4497ms 2.3486ms 425.7817 Ops/s 384.0401 Ops/s $\textbf{\color{#35bf28}+10.87\%}$
test_ddpg_speed[reduce-overhead-None] 1.7713ms 1.3815ms 723.8329 Ops/s 707.3817 Ops/s $\color{#35bf28}+2.33\%$
test_ddpg_speed[reduce-overhead-backward] 2.7267ms 2.3457ms 426.3120 Ops/s 439.1455 Ops/s $\color{#d91a1a}-2.92\%$
test_sac_speed[False-None] 8.1267ms 7.6698ms 130.3813 Ops/s 98.1099 Ops/s $\textbf{\color{#35bf28}+32.89\%}$
test_sac_speed[False-backward] 11.2117ms 10.8339ms 92.3026 Ops/s 96.5537 Ops/s $\color{#d91a1a}-4.40\%$
test_sac_speed[True-None] 2.4739ms 2.0930ms 477.7807 Ops/s 458.1816 Ops/s $\color{#35bf28}+4.28\%$
test_sac_speed[True-backward] 4.2995ms 3.9741ms 251.6270 Ops/s 252.4788 Ops/s $\color{#d91a1a}-0.34\%$
test_sac_speed[reduce-overhead-None] 2.5621ms 2.0881ms 478.9014 Ops/s 456.8142 Ops/s $\color{#35bf28}+4.84\%$
test_sac_speed[reduce-overhead-backward] 4.1156ms 3.9782ms 251.3706 Ops/s 251.7723 Ops/s $\color{#d91a1a}-0.16\%$
test_redq_speed[False-None] 10.4637ms 9.9782ms 100.2184 Ops/s 100.3476 Ops/s $\color{#d91a1a}-0.13\%$
test_redq_speed[False-backward] 18.1774ms 17.4137ms 57.4262 Ops/s 59.7198 Ops/s $\color{#d91a1a}-3.84\%$
test_redq_speed[True-None] 4.5213ms 4.2767ms 233.8275 Ops/s 217.6083 Ops/s $\textbf{\color{#35bf28}+7.45\%}$
test_redq_speed[True-backward] 9.9952ms 9.5841ms 104.3400 Ops/s 103.6878 Ops/s $\color{#35bf28}+0.63\%$
test_redq_speed[reduce-overhead-None] 5.3220ms 4.2995ms 232.5841 Ops/s 218.8467 Ops/s $\textbf{\color{#35bf28}+6.28\%}$
test_redq_speed[reduce-overhead-backward] 10.0377ms 9.7910ms 102.1343 Ops/s 95.1315 Ops/s $\textbf{\color{#35bf28}+7.36\%}$
test_redq_deprec_speed[False-None] 11.1455ms 10.6551ms 93.8520 Ops/s 93.3734 Ops/s $\color{#35bf28}+0.51\%$
test_redq_deprec_speed[False-backward] 15.6906ms 15.3185ms 65.2805 Ops/s 66.1866 Ops/s $\color{#d91a1a}-1.37\%$
test_redq_deprec_speed[True-None] 3.8212ms 3.5481ms 281.8390 Ops/s 277.4548 Ops/s $\color{#35bf28}+1.58\%$
test_redq_deprec_speed[True-backward] 7.4734ms 7.2066ms 138.7612 Ops/s 130.9055 Ops/s $\textbf{\color{#35bf28}+6.00\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8190ms 3.4943ms 286.1818 Ops/s 280.7897 Ops/s $\color{#35bf28}+1.92\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.3954ms 7.1895ms 139.0922 Ops/s 126.8760 Ops/s $\textbf{\color{#35bf28}+9.63\%}$
test_td3_speed[False-None] 49.8825ms 8.1358ms 122.9140 Ops/s 132.1126 Ops/s $\textbf{\color{#d91a1a}-6.96\%}$
test_td3_speed[False-backward] 11.0145ms 10.5036ms 95.2058 Ops/s 98.8626 Ops/s $\color{#d91a1a}-3.70\%$
test_td3_speed[True-None] 1.8913ms 1.8338ms 545.3167 Ops/s 544.6600 Ops/s $\color{#35bf28}+0.12\%$
test_td3_speed[True-backward] 3.7783ms 3.5946ms 278.1961 Ops/s 244.3962 Ops/s $\textbf{\color{#35bf28}+13.83\%}$
test_td3_speed[reduce-overhead-None] 1.8126ms 1.7613ms 567.7679 Ops/s 559.0430 Ops/s $\color{#35bf28}+1.56\%$
test_td3_speed[reduce-overhead-backward] 3.7497ms 3.5735ms 279.8367 Ops/s 229.2782 Ops/s $\textbf{\color{#35bf28}+22.05\%}$
test_cql_speed[False-None] 25.4979ms 24.9351ms 40.1041 Ops/s 39.2736 Ops/s $\color{#35bf28}+2.11\%$
test_cql_speed[False-backward] 34.4123ms 33.8264ms 29.5627 Ops/s 29.7363 Ops/s $\color{#d91a1a}-0.58\%$
test_cql_speed[True-None] 12.4860ms 12.2969ms 81.3216 Ops/s 80.5860 Ops/s $\color{#35bf28}+0.91\%$
test_cql_speed[True-backward] 18.4094ms 17.7941ms 56.1983 Ops/s 55.5451 Ops/s $\color{#35bf28}+1.18\%$
test_cql_speed[reduce-overhead-None] 12.6610ms 12.3801ms 80.7745 Ops/s 79.6460 Ops/s $\color{#35bf28}+1.42\%$
test_cql_speed[reduce-overhead-backward] 18.2383ms 17.7120ms 56.4589 Ops/s 52.4794 Ops/s $\textbf{\color{#35bf28}+7.58\%}$
test_a2c_speed[False-None] 5.6236ms 5.3463ms 187.0440 Ops/s 183.0009 Ops/s $\color{#35bf28}+2.21\%$
test_a2c_speed[False-backward] 12.0040ms 11.6569ms 85.7862 Ops/s 85.1554 Ops/s $\color{#35bf28}+0.74\%$
test_a2c_speed[True-None] 4.0520ms 3.7144ms 269.2197 Ops/s 257.4899 Ops/s $\color{#35bf28}+4.56\%$
test_a2c_speed[True-backward] 8.8435ms 8.5254ms 117.2964 Ops/s 116.1268 Ops/s $\color{#35bf28}+1.01\%$
test_a2c_speed[reduce-overhead-None] 3.8814ms 3.6985ms 270.3774 Ops/s 270.9078 Ops/s $\color{#d91a1a}-0.20\%$
test_a2c_speed[reduce-overhead-backward] 9.0276ms 8.8523ms 112.9650 Ops/s 110.7843 Ops/s $\color{#35bf28}+1.97\%$
test_ppo_speed[False-None] 6.2307ms 5.7738ms 173.1961 Ops/s 172.4595 Ops/s $\color{#35bf28}+0.43\%$
test_ppo_speed[False-backward] 12.7233ms 12.3924ms 80.6949 Ops/s 82.5379 Ops/s $\color{#d91a1a}-2.23\%$
test_ppo_speed[True-None] 3.8136ms 3.6655ms 272.8117 Ops/s 268.6523 Ops/s $\color{#35bf28}+1.55\%$
test_ppo_speed[True-backward] 9.2138ms 8.4890ms 117.7989 Ops/s 118.4472 Ops/s $\color{#d91a1a}-0.55\%$
test_ppo_speed[reduce-overhead-None] 3.7678ms 3.6260ms 275.7848 Ops/s 271.1937 Ops/s $\color{#35bf28}+1.69\%$
test_ppo_speed[reduce-overhead-backward] 9.1605ms 8.8211ms 113.3647 Ops/s 114.7400 Ops/s $\color{#d91a1a}-1.20\%$
test_reinforce_speed[False-None] 4.7248ms 4.5146ms 221.5060 Ops/s 218.0367 Ops/s $\color{#35bf28}+1.59\%$
test_reinforce_speed[False-backward] 7.5655ms 7.2986ms 137.0127 Ops/s 137.0711 Ops/s $\color{#d91a1a}-0.04\%$
test_reinforce_speed[True-None] 2.9821ms 2.8584ms 349.8462 Ops/s 340.3623 Ops/s $\color{#35bf28}+2.79\%$
test_reinforce_speed[True-backward] 7.8711ms 7.7065ms 129.7599 Ops/s 129.4067 Ops/s $\color{#35bf28}+0.27\%$
test_reinforce_speed[reduce-overhead-None] 3.0919ms 2.8387ms 352.2755 Ops/s 348.0377 Ops/s $\color{#35bf28}+1.22\%$
test_reinforce_speed[reduce-overhead-backward] 8.2309ms 7.8948ms 126.6653 Ops/s 127.1658 Ops/s $\color{#d91a1a}-0.39\%$
test_iql_speed[False-None] 20.2770ms 19.5887ms 51.0497 Ops/s 51.5156 Ops/s $\color{#d91a1a}-0.90\%$
test_iql_speed[False-backward] 30.9739ms 30.1153ms 33.2058 Ops/s 34.0148 Ops/s $\color{#d91a1a}-2.38\%$
test_iql_speed[True-None] 8.7477ms 8.5092ms 117.5194 Ops/s 116.7528 Ops/s $\color{#35bf28}+0.66\%$
test_iql_speed[True-backward] 17.2419ms 16.7166ms 59.8209 Ops/s 60.1123 Ops/s $\color{#d91a1a}-0.48\%$
test_iql_speed[reduce-overhead-None] 8.9722ms 8.6404ms 115.7358 Ops/s 115.9384 Ops/s $\color{#d91a1a}-0.17\%$
test_iql_speed[reduce-overhead-backward] 17.9388ms 17.2241ms 58.0581 Ops/s 58.3289 Ops/s $\color{#d91a1a}-0.46\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4658ms 5.9878ms 167.0072 Ops/s 173.9072 Ops/s $\color{#d91a1a}-3.97\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5265ms 0.2748ms 3.6391 KOps/s 2.9293 KOps/s $\textbf{\color{#35bf28}+24.23\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5492ms 0.2554ms 3.9149 KOps/s 3.0765 KOps/s $\textbf{\color{#35bf28}+27.25\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9920ms 5.7561ms 173.7288 Ops/s 181.5816 Ops/s $\color{#d91a1a}-4.32\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9529ms 0.2718ms 3.6788 KOps/s 2.8374 KOps/s $\textbf{\color{#35bf28}+29.65\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5062ms 0.2529ms 3.9537 KOps/s 3.0878 KOps/s $\textbf{\color{#35bf28}+28.04\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4427ms 1.2078ms 827.9684 Ops/s 746.7748 Ops/s $\textbf{\color{#35bf28}+10.87\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.3248ms 1.1232ms 890.2843 Ops/s 795.4779 Ops/s $\textbf{\color{#35bf28}+11.92\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.0551ms 6.0744ms 164.6255 Ops/s 175.0055 Ops/s $\textbf{\color{#d91a1a}-5.93\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1191ms 0.4786ms 2.0896 KOps/s 1.9760 KOps/s $\textbf{\color{#35bf28}+5.75\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7670ms 0.4827ms 2.0717 KOps/s 2.0505 KOps/s $\color{#35bf28}+1.04\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1206ms 5.8145ms 171.9840 Ops/s 179.8129 Ops/s $\color{#d91a1a}-4.35\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0475ms 0.2811ms 3.5579 KOps/s 2.5877 KOps/s $\textbf{\color{#35bf28}+37.49\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4609ms 0.2612ms 3.8285 KOps/s 2.9100 KOps/s $\textbf{\color{#35bf28}+31.56\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9762ms 5.7458ms 174.0407 Ops/s 180.0681 Ops/s $\color{#d91a1a}-3.35\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9851ms 0.3449ms 2.8997 KOps/s 3.3648 KOps/s $\textbf{\color{#d91a1a}-13.82\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4744ms 0.2577ms 3.8799 KOps/s 3.9672 KOps/s $\color{#d91a1a}-2.20\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.9641ms 5.8854ms 169.9107 Ops/s 174.1490 Ops/s $\color{#d91a1a}-2.43\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1893ms 0.4263ms 2.3456 KOps/s 2.1981 KOps/s $\textbf{\color{#35bf28}+6.71\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6327ms 0.4063ms 2.4614 KOps/s 2.2081 KOps/s $\textbf{\color{#35bf28}+11.47\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.6426ms 5.1210ms 195.2739 Ops/s 54.6879 Ops/s $\textbf{\color{#35bf28}+257.07\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.6291ms 2.3063ms 433.5911 Ops/s 503.2951 Ops/s $\textbf{\color{#d91a1a}-13.85\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.8674ms 1.1733ms 852.2781 Ops/s 823.7733 Ops/s $\color{#35bf28}+3.46\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5483s 15.9426ms 62.7251 Ops/s 201.9572 Ops/s $\textbf{\color{#d91a1a}-68.94\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.9127ms 1.7312ms 577.6246 Ops/s 509.1745 Ops/s $\textbf{\color{#35bf28}+13.44\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.1988ms 1.0546ms 948.2400 Ops/s 829.5255 Ops/s $\textbf{\color{#35bf28}+14.31\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.3948ms 5.2777ms 189.4750 Ops/s 195.6707 Ops/s $\color{#d91a1a}-3.17\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.2844ms 2.2121ms 452.0612 Ops/s 455.6822 Ops/s $\color{#d91a1a}-0.79\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.8560ms 1.3815ms 723.8317 Ops/s 741.4657 Ops/s $\color{#d91a1a}-2.38\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 34.2828ms 32.0448ms 31.2064 Ops/s 30.6434 Ops/s $\color{#35bf28}+1.84\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.7044ms 16.9680ms 58.9345 Ops/s 58.2293 Ops/s $\color{#35bf28}+1.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 36.1909ms 33.1883ms 30.1311 Ops/s 29.5321 Ops/s $\color{#35bf28}+2.03\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.6281ms 17.5306ms 57.0430 Ops/s 58.2540 Ops/s $\color{#d91a1a}-2.08\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 37.4408ms 35.5891ms 28.0985 Ops/s 28.1588 Ops/s $\color{#d91a1a}-0.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 0.4544s 27.7794ms 35.9979 Ops/s 53.7721 Ops/s $\textbf{\color{#d91a1a}-33.05\%}$

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 20, 2025
@vmoens vmoens merged commit 0516176 into gh/vmoens/148/base Oct 20, 2025
70 of 82 checks passed
@vmoens vmoens deleted the gh/vmoens/148/head branch October 20, 2025 00:09
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