+
Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Jul 28, 2025

No description provided.

Copy link

pytorch-bot bot commented Jul 28, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3099

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

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

@facebook-github-bot facebook-github-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 Jul 28, 2025
@vmoens vmoens added the BE Better errors, logs, docs or test utils label Jul 28, 2025
@vmoens vmoens force-pushed the fix-annoying-warning branch from 5d9e285 to 7d2ebbb Compare July 28, 2025 16:26
@vmoens
Copy link
Collaborator Author

vmoens commented Jul 28, 2025

Tested locally and minor cosmetic changes - merging

@vmoens vmoens merged commit c226646 into main Jul 28, 2025
52 of 67 checks passed
@vmoens vmoens deleted the fix-annoying-warning branch July 28, 2025 16:34
Copy link

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.4701μs 80.9439μs 12.3542 KOps/s 12.2224 KOps/s $\color{#35bf28}+1.08\%$
test_tensor_to_bytestream_speed[torch.save] 0.1419ms 0.1403ms 7.1257 KOps/s 6.7896 KOps/s $\color{#35bf28}+4.95\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1091s 0.1088s 9.1871 Ops/s 9.1303 Ops/s $\color{#35bf28}+0.62\%$
test_tensor_to_bytestream_speed[numpy] 2.7864μs 2.7757μs 360.2719 KOps/s 362.1252 KOps/s $\color{#d91a1a}-0.51\%$
test_tensor_to_bytestream_speed[safetensors] 40.3191μs 40.2016μs 24.8746 KOps/s 24.2218 KOps/s $\color{#35bf28}+2.70\%$
test_simple 0.5419s 0.5413s 1.8474 Ops/s 1.8360 Ops/s $\color{#35bf28}+0.62\%$
test_transformed 1.1097s 1.1077s 0.9028 Ops/s 0.8830 Ops/s $\color{#35bf28}+2.24\%$
test_serial 1.7611s 1.6766s 0.5965 Ops/s 0.5936 Ops/s $\color{#35bf28}+0.49\%$
test_parallel 1.1652s 1.0790s 0.9268 Ops/s 0.9251 Ops/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[True-True-True-True-True] 0.1442ms 44.7806μs 22.3311 KOps/s 21.9754 KOps/s $\color{#35bf28}+1.62\%$
test_step_mdp_speed[True-True-True-True-False] 59.8940μs 25.4135μs 39.3492 KOps/s 39.5224 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[True-True-True-False-True] 54.9340μs 25.1378μs 39.7808 KOps/s 39.1540 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[True-True-True-False-False] 49.6330μs 13.8443μs 72.2319 KOps/s 72.3249 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-True-False-True-True] 0.1123ms 47.6222μs 20.9986 KOps/s 20.9869 KOps/s $\color{#35bf28}+0.06\%$
test_step_mdp_speed[True-True-False-True-False] 63.8250μs 28.2847μs 35.3548 KOps/s 35.8146 KOps/s $\color{#d91a1a}-1.28\%$
test_step_mdp_speed[True-True-False-False-True] 77.8350μs 28.1800μs 35.4862 KOps/s 35.9054 KOps/s $\color{#d91a1a}-1.17\%$
test_step_mdp_speed[True-True-False-False-False] 59.6440μs 16.7263μs 59.7862 KOps/s 60.0698 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-False-True-True-True] 0.1087ms 50.9998μs 19.6079 KOps/s 19.7061 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-False-True-True-False] 69.6950μs 30.9548μs 32.3052 KOps/s 32.1190 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[True-False-True-False-True] 57.7340μs 28.6188μs 34.9420 KOps/s 35.5238 KOps/s $\color{#d91a1a}-1.64\%$
test_step_mdp_speed[True-False-True-False-False] 46.9830μs 16.6621μs 60.0165 KOps/s 60.3096 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[True-False-False-True-True] 0.2433ms 53.6094μs 18.6535 KOps/s 19.0733 KOps/s $\color{#d91a1a}-2.20\%$
test_step_mdp_speed[True-False-False-True-False] 64.4550μs 33.7529μs 29.6271 KOps/s 30.2256 KOps/s $\color{#d91a1a}-1.98\%$
test_step_mdp_speed[True-False-False-False-True] 0.1790ms 30.7793μs 32.4893 KOps/s 32.8728 KOps/s $\color{#d91a1a}-1.17\%$
test_step_mdp_speed[True-False-False-False-False] 49.9230μs 19.5995μs 51.0218 KOps/s 51.8436 KOps/s $\color{#d91a1a}-1.59\%$
test_step_mdp_speed[False-True-True-True-True] 84.5560μs 50.0678μs 19.9729 KOps/s 19.9807 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[False-True-True-True-False] 59.9640μs 31.4369μs 31.8097 KOps/s 32.6112 KOps/s $\color{#d91a1a}-2.46\%$
test_step_mdp_speed[False-True-True-False-True] 59.8240μs 31.8603μs 31.3871 KOps/s 32.4678 KOps/s $\color{#d91a1a}-3.33\%$
test_step_mdp_speed[False-True-True-False-False] 52.7740μs 19.1461μs 52.2300 KOps/s 52.5078 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[False-True-False-True-True] 2.6665ms 54.2803μs 18.4229 KOps/s 18.7348 KOps/s $\color{#d91a1a}-1.67\%$
test_step_mdp_speed[False-True-False-True-False] 81.6460μs 34.4115μs 29.0600 KOps/s 29.8652 KOps/s $\color{#d91a1a}-2.70\%$
test_step_mdp_speed[False-True-False-False-True] 74.5550μs 35.3918μs 28.2551 KOps/s 29.0445 KOps/s $\color{#d91a1a}-2.72\%$
test_step_mdp_speed[False-True-False-False-False] 59.8840μs 21.8178μs 45.8340 KOps/s 46.2309 KOps/s $\color{#d91a1a}-0.86\%$
test_step_mdp_speed[False-False-True-True-True] 0.1048ms 56.0408μs 17.8442 KOps/s 18.2307 KOps/s $\color{#d91a1a}-2.12\%$
test_step_mdp_speed[False-False-True-True-False] 73.0550μs 36.5355μs 27.3706 KOps/s 27.7186 KOps/s $\color{#d91a1a}-1.26\%$
test_step_mdp_speed[False-False-True-False-True] 0.1250ms 35.0472μs 28.5329 KOps/s 29.2257 KOps/s $\color{#d91a1a}-2.37\%$
test_step_mdp_speed[False-False-True-False-False] 50.8940μs 21.5177μs 46.4734 KOps/s 46.0603 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[False-False-False-True-True] 90.6860μs 58.3574μs 17.1358 KOps/s 17.3905 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[False-False-False-True-False] 0.1273ms 38.4834μs 25.9852 KOps/s 25.9530 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[False-False-False-False-True] 77.3750μs 37.2261μs 26.8629 KOps/s 27.8347 KOps/s $\color{#d91a1a}-3.49\%$
test_step_mdp_speed[False-False-False-False-False] 47.3330μs 24.0236μs 41.6258 KOps/s 42.7935 KOps/s $\color{#d91a1a}-2.73\%$
test_values[generalized_advantage_estimate-True-True] 10.9359ms 10.6827ms 93.6091 Ops/s 94.1410 Ops/s $\color{#d91a1a}-0.57\%$
test_values[vec_generalized_advantage_estimate-True-True] 15.2359ms 11.3787ms 87.8834 Ops/s 90.1277 Ops/s $\color{#d91a1a}-2.49\%$
test_values[td0_return_estimate-False-False] 0.2166ms 0.1294ms 7.7257 KOps/s 7.9105 KOps/s $\color{#d91a1a}-2.34\%$
test_values[td1_return_estimate-False-False] 28.5662ms 27.8861ms 35.8601 Ops/s 36.0584 Ops/s $\color{#d91a1a}-0.55\%$
test_values[vec_td1_return_estimate-False-False] 11.5698ms 11.0396ms 90.5829 Ops/s 89.8682 Ops/s $\color{#35bf28}+0.80\%$
test_values[td_lambda_return_estimate-True-False] 42.2612ms 41.1625ms 24.2940 Ops/s 24.1057 Ops/s $\color{#35bf28}+0.78\%$
test_values[vec_td_lambda_return_estimate-True-False] 13.0225ms 11.0683ms 90.3481 Ops/s 90.6011 Ops/s $\color{#d91a1a}-0.28\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.4024ms 9.3103ms 107.4082 Ops/s 108.0718 Ops/s $\color{#d91a1a}-0.61\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7737ms 1.5227ms 656.7189 Ops/s 650.7553 Ops/s $\color{#35bf28}+0.92\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5758ms 0.4168ms 2.3995 KOps/s 2.4433 KOps/s $\color{#d91a1a}-1.79\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 30.0004ms 24.2720ms 41.1998 Ops/s 34.1197 Ops/s $\textbf{\color{#35bf28}+20.75\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8725ms 1.7057ms 586.2565 Ops/s 578.3839 Ops/s $\color{#35bf28}+1.36\%$
test_dqn_speed[False-None] 1.7720ms 1.3622ms 734.1035 Ops/s 725.1641 Ops/s $\color{#35bf28}+1.23\%$
test_dqn_speed[False-backward] 2.0535ms 1.8951ms 527.6834 Ops/s 533.9755 Ops/s $\color{#d91a1a}-1.18\%$
test_dqn_speed[True-None] 0.6966ms 0.5151ms 1.9412 KOps/s 1.9164 KOps/s $\color{#35bf28}+1.30\%$
test_dqn_speed[True-backward] 0.9892ms 0.9617ms 1.0398 KOps/s 864.8309 Ops/s $\textbf{\color{#35bf28}+20.23\%}$
test_dqn_speed[reduce-overhead-None] 0.6583ms 0.5158ms 1.9388 KOps/s 1.8200 KOps/s $\textbf{\color{#35bf28}+6.53\%}$
test_dqn_speed[reduce-overhead-backward] 1.1323ms 0.9710ms 1.0298 KOps/s 1.0053 KOps/s $\color{#35bf28}+2.44\%$
test_ddpg_speed[False-None] 3.1010ms 2.8212ms 354.4561 Ops/s 352.8385 Ops/s $\color{#35bf28}+0.46\%$
test_ddpg_speed[False-backward] 4.0705ms 3.9917ms 250.5201 Ops/s 249.3686 Ops/s $\color{#35bf28}+0.46\%$
test_ddpg_speed[True-None] 1.5425ms 1.3695ms 730.1767 Ops/s 713.1466 Ops/s $\color{#35bf28}+2.39\%$
test_ddpg_speed[True-backward] 2.4073ms 2.3643ms 422.9578 Ops/s 413.5352 Ops/s $\color{#35bf28}+2.28\%$
test_ddpg_speed[reduce-overhead-None] 1.5454ms 1.3721ms 728.8130 Ops/s 716.6631 Ops/s $\color{#35bf28}+1.70\%$
test_ddpg_speed[reduce-overhead-backward] 2.5519ms 2.3737ms 421.2840 Ops/s 422.2243 Ops/s $\color{#d91a1a}-0.22\%$
test_sac_speed[False-None] 8.3936ms 7.7030ms 129.8199 Ops/s 130.6303 Ops/s $\color{#d91a1a}-0.62\%$
test_sac_speed[False-backward] 11.1642ms 10.7815ms 92.7513 Ops/s 92.0641 Ops/s $\color{#35bf28}+0.75\%$
test_sac_speed[True-None] 2.2676ms 2.0962ms 477.0492 Ops/s 455.9626 Ops/s $\color{#35bf28}+4.62\%$
test_sac_speed[True-backward] 4.1096ms 3.9575ms 252.6823 Ops/s 219.8155 Ops/s $\textbf{\color{#35bf28}+14.95\%}$
test_sac_speed[reduce-overhead-None] 2.2981ms 2.1261ms 470.3382 Ops/s 459.2899 Ops/s $\color{#35bf28}+2.41\%$
test_sac_speed[reduce-overhead-backward] 4.1111ms 4.0062ms 249.6145 Ops/s 244.6607 Ops/s $\color{#35bf28}+2.02\%$
test_redq_speed[False-None] 10.5478ms 9.9623ms 100.3780 Ops/s 96.4927 Ops/s $\color{#35bf28}+4.03\%$
test_redq_speed[False-backward] 18.0172ms 17.2988ms 57.8073 Ops/s 55.8518 Ops/s $\color{#35bf28}+3.50\%$
test_redq_speed[True-None] 4.4661ms 4.2069ms 237.7034 Ops/s 228.2206 Ops/s $\color{#35bf28}+4.16\%$
test_redq_speed[True-backward] 9.6073ms 9.3054ms 107.4639 Ops/s 103.0495 Ops/s $\color{#35bf28}+4.28\%$
test_redq_speed[reduce-overhead-None] 4.4978ms 4.2694ms 234.2236 Ops/s 233.4129 Ops/s $\color{#35bf28}+0.35\%$
test_redq_speed[reduce-overhead-backward] 9.7261ms 9.3193ms 107.3046 Ops/s 102.9704 Ops/s $\color{#35bf28}+4.21\%$
test_redq_deprec_speed[False-None] 11.0145ms 10.4936ms 95.2966 Ops/s 94.0916 Ops/s $\color{#35bf28}+1.28\%$
test_redq_deprec_speed[False-backward] 15.6691ms 15.1194ms 66.1402 Ops/s 65.2050 Ops/s $\color{#35bf28}+1.43\%$
test_redq_deprec_speed[True-None] 3.8604ms 3.4663ms 288.4938 Ops/s 280.9121 Ops/s $\color{#35bf28}+2.70\%$
test_redq_deprec_speed[True-backward] 7.4083ms 7.0353ms 142.1398 Ops/s 117.6390 Ops/s $\textbf{\color{#35bf28}+20.83\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.6808ms 3.4432ms 290.4249 Ops/s 265.1791 Ops/s $\textbf{\color{#35bf28}+9.52\%}$
test_redq_deprec_speed[reduce-overhead-backward] 7.3770ms 7.0773ms 141.2960 Ops/s 122.2552 Ops/s $\textbf{\color{#35bf28}+15.57\%}$
test_td3_speed[False-None] 7.8689ms 7.6551ms 130.6319 Ops/s 128.7247 Ops/s $\color{#35bf28}+1.48\%$
test_td3_speed[False-backward] 11.0178ms 10.5133ms 95.1175 Ops/s 94.9946 Ops/s $\color{#35bf28}+0.13\%$
test_td3_speed[True-None] 1.8208ms 1.7880ms 559.2940 Ops/s 538.5594 Ops/s $\color{#35bf28}+3.85\%$
test_td3_speed[True-backward] 3.5498ms 3.4736ms 287.8819 Ops/s 281.1552 Ops/s $\color{#35bf28}+2.39\%$
test_td3_speed[reduce-overhead-None] 1.8255ms 1.7857ms 559.9899 Ops/s 545.3910 Ops/s $\color{#35bf28}+2.68\%$
test_td3_speed[reduce-overhead-backward] 3.5970ms 3.5066ms 285.1782 Ops/s 280.0262 Ops/s $\color{#35bf28}+1.84\%$
test_cql_speed[False-None] 28.5975ms 25.4218ms 39.3364 Ops/s 39.1039 Ops/s $\color{#35bf28}+0.59\%$
test_cql_speed[False-backward] 35.2818ms 34.3480ms 29.1138 Ops/s 28.2156 Ops/s $\color{#35bf28}+3.18\%$
test_cql_speed[True-None] 12.2493ms 11.9542ms 83.6526 Ops/s 82.8825 Ops/s $\color{#35bf28}+0.93\%$
test_cql_speed[True-backward] 17.7274ms 17.4387ms 57.3439 Ops/s 54.3842 Ops/s $\textbf{\color{#35bf28}+5.44\%}$
test_cql_speed[reduce-overhead-None] 12.3214ms 12.0840ms 82.7538 Ops/s 81.2346 Ops/s $\color{#35bf28}+1.87\%$
test_cql_speed[reduce-overhead-backward] 18.0280ms 17.5003ms 57.1419 Ops/s 54.8406 Ops/s $\color{#35bf28}+4.20\%$
test_a2c_speed[False-None] 5.6596ms 5.3246ms 187.8087 Ops/s 182.5838 Ops/s $\color{#35bf28}+2.86\%$
test_a2c_speed[False-backward] 12.0699ms 11.7537ms 85.0795 Ops/s 84.9679 Ops/s $\color{#35bf28}+0.13\%$
test_a2c_speed[True-None] 3.9179ms 3.7040ms 269.9774 Ops/s 270.0493 Ops/s $\color{#d91a1a}-0.03\%$
test_a2c_speed[True-backward] 9.0594ms 8.6281ms 115.9008 Ops/s 114.5972 Ops/s $\color{#35bf28}+1.14\%$
test_a2c_speed[reduce-overhead-None] 3.8899ms 3.6840ms 271.4442 Ops/s 271.4805 Ops/s $\color{#d91a1a}-0.01\%$
test_a2c_speed[reduce-overhead-backward] 8.8999ms 8.5166ms 117.4180 Ops/s 117.0068 Ops/s $\color{#35bf28}+0.35\%$
test_ppo_speed[False-None] 6.0235ms 5.7432ms 174.1179 Ops/s 176.3776 Ops/s $\color{#d91a1a}-1.28\%$
test_ppo_speed[False-backward] 13.1289ms 12.3785ms 80.7851 Ops/s 82.3596 Ops/s $\color{#d91a1a}-1.91\%$
test_ppo_speed[True-None] 3.8432ms 3.6164ms 276.5152 Ops/s 269.5993 Ops/s $\color{#35bf28}+2.57\%$
test_ppo_speed[True-backward] 8.7821ms 8.4496ms 118.3484 Ops/s 117.0950 Ops/s $\color{#35bf28}+1.07\%$
test_ppo_speed[reduce-overhead-None] 3.7985ms 3.5801ms 279.3250 Ops/s 271.3362 Ops/s $\color{#35bf28}+2.94\%$
test_ppo_speed[reduce-overhead-backward] 8.8472ms 8.4232ms 118.7198 Ops/s 118.4527 Ops/s $\color{#35bf28}+0.23\%$
test_reinforce_speed[False-None] 4.9745ms 4.4670ms 223.8632 Ops/s 218.5473 Ops/s $\color{#35bf28}+2.43\%$
test_reinforce_speed[False-backward] 7.5718ms 7.3633ms 135.8083 Ops/s 135.9757 Ops/s $\color{#d91a1a}-0.12\%$
test_reinforce_speed[True-None] 3.1303ms 2.8537ms 350.4218 Ops/s 350.1304 Ops/s $\color{#35bf28}+0.08\%$
test_reinforce_speed[True-backward] 7.7096ms 7.4952ms 133.4186 Ops/s 125.9745 Ops/s $\textbf{\color{#35bf28}+5.91\%}$
test_reinforce_speed[reduce-overhead-None] 3.0507ms 2.8254ms 353.9267 Ops/s 345.4952 Ops/s $\color{#35bf28}+2.44\%$
test_reinforce_speed[reduce-overhead-backward] 7.8422ms 7.5423ms 132.5859 Ops/s 128.7777 Ops/s $\color{#35bf28}+2.96\%$
test_iql_speed[False-None] 20.1494ms 19.3841ms 51.5886 Ops/s 49.5487 Ops/s $\color{#35bf28}+4.12\%$
test_iql_speed[False-backward] 30.4343ms 29.7456ms 33.6184 Ops/s 32.9052 Ops/s $\color{#35bf28}+2.17\%$
test_iql_speed[True-None] 8.5672ms 8.3150ms 120.2651 Ops/s 114.6687 Ops/s $\color{#35bf28}+4.88\%$
test_iql_speed[True-backward] 17.2185ms 16.4280ms 60.8716 Ops/s 59.5377 Ops/s $\color{#35bf28}+2.24\%$
test_iql_speed[reduce-overhead-None] 8.6669ms 8.4106ms 118.8981 Ops/s 117.8067 Ops/s $\color{#35bf28}+0.93\%$
test_iql_speed[reduce-overhead-backward] 16.7949ms 16.3452ms 61.1799 Ops/s 59.4956 Ops/s $\color{#35bf28}+2.83\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5554ms 6.1670ms 162.1522 Ops/s 161.0378 Ops/s $\color{#35bf28}+0.69\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5839ms 0.3490ms 2.8651 KOps/s 3.7285 KOps/s $\textbf{\color{#d91a1a}-23.16\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6412ms 0.3364ms 2.9723 KOps/s 4.0403 KOps/s $\textbf{\color{#d91a1a}-26.43\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2198ms 5.9180ms 168.9769 Ops/s 169.1256 Ops/s $\color{#d91a1a}-0.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1630ms 0.3443ms 2.9041 KOps/s 3.7779 KOps/s $\textbf{\color{#d91a1a}-23.13\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5676ms 0.3214ms 3.1112 KOps/s 3.2111 KOps/s $\color{#d91a1a}-3.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6449ms 1.3564ms 737.2527 Ops/s 722.3552 Ops/s $\color{#35bf28}+2.06\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5219ms 1.2847ms 778.3918 Ops/s 798.4908 Ops/s $\color{#d91a1a}-2.52\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2192ms 6.0465ms 165.3852 Ops/s 164.2813 Ops/s $\color{#35bf28}+0.67\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.6917ms 0.4407ms 2.2693 KOps/s 2.1060 KOps/s $\textbf{\color{#35bf28}+7.75\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6934ms 0.4244ms 2.3563 KOps/s 2.3109 KOps/s $\color{#35bf28}+1.97\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1518ms 5.9734ms 167.4100 Ops/s 166.2908 Ops/s $\color{#35bf28}+0.67\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.7980ms 0.2990ms 3.3447 KOps/s 3.7154 KOps/s $\textbf{\color{#d91a1a}-9.98\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5495ms 0.2957ms 3.3813 KOps/s 4.0540 KOps/s $\textbf{\color{#d91a1a}-16.59\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.4779ms 5.8751ms 170.2105 Ops/s 167.6207 Ops/s $\color{#35bf28}+1.55\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6749ms 0.3466ms 2.8848 KOps/s 3.6456 KOps/s $\textbf{\color{#d91a1a}-20.87\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6363ms 0.3021ms 3.3105 KOps/s 3.5570 KOps/s $\textbf{\color{#d91a1a}-6.93\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2620ms 6.0804ms 164.4625 Ops/s 163.2159 Ops/s $\color{#35bf28}+0.76\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.0818ms 0.4931ms 2.0281 KOps/s 2.1960 KOps/s $\textbf{\color{#d91a1a}-7.64\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6810ms 0.4563ms 2.1914 KOps/s 2.4145 KOps/s $\textbf{\color{#d91a1a}-9.24\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1285ms 5.5697ms 179.5425 Ops/s 177.9732 Ops/s $\color{#35bf28}+0.88\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.7694ms 2.1089ms 474.1797 Ops/s 433.6537 Ops/s $\textbf{\color{#35bf28}+9.35\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.1841ms 1.3145ms 760.7646 Ops/s 763.8522 Ops/s $\color{#d91a1a}-0.40\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4313s 14.1131ms 70.8561 Ops/s 60.3570 Ops/s $\textbf{\color{#35bf28}+17.39\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.3860ms 2.0954ms 477.2302 Ops/s 488.3254 Ops/s $\color{#d91a1a}-2.27\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.0129ms 1.2556ms 796.4122 Ops/s 775.1413 Ops/s $\color{#35bf28}+2.74\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.4937ms 5.7996ms 172.4267 Ops/s 171.3046 Ops/s $\color{#35bf28}+0.66\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.1747ms 2.2963ms 435.4917 Ops/s 443.3290 Ops/s $\color{#d91a1a}-1.77\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.7831ms 1.4427ms 693.1586 Ops/s 697.5077 Ops/s $\color{#d91a1a}-0.62\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 59.4062ms 57.6208ms 17.3548 Ops/s 17.1694 Ops/s $\color{#35bf28}+1.08\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.2954ms 16.7514ms 59.6963 Ops/s 60.5249 Ops/s $\color{#d91a1a}-1.37\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 62.3424ms 58.9785ms 16.9553 Ops/s 17.2063 Ops/s $\color{#d91a1a}-1.46\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7697ms 16.8534ms 59.3351 Ops/s 59.6674 Ops/s $\color{#d91a1a}-0.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 61.8141ms 60.2536ms 16.5965 Ops/s 17.0188 Ops/s $\color{#d91a1a}-2.48\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.3279ms 18.8507ms 53.0483 Ops/s 53.5666 Ops/s $\color{#d91a1a}-0.97\%$

Copy link

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

Total Benchmarks: 148. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 79.5893μs 78.5239μs 12.7350 KOps/s 12.1446 KOps/s $\color{#35bf28}+4.86\%$
test_tensor_to_bytestream_speed[torch.save] 0.1432ms 0.1398ms 7.1513 KOps/s 6.9225 KOps/s $\color{#35bf28}+3.30\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1066s 0.1060s 9.4365 Ops/s 8.9831 Ops/s $\textbf{\color{#35bf28}+5.05\%}$
test_tensor_to_bytestream_speed[numpy] 2.7704μs 2.7642μs 361.7627 KOps/s 353.8189 KOps/s $\color{#35bf28}+2.25\%$
test_tensor_to_bytestream_speed[safetensors] 42.0846μs 41.9026μs 23.8649 KOps/s 23.7752 KOps/s $\color{#35bf28}+0.38\%$
test_simple 0.7642s 0.7604s 1.3150 Ops/s 1.2763 Ops/s $\color{#35bf28}+3.03\%$
test_transformed 1.3638s 1.3628s 0.7338 Ops/s 0.7235 Ops/s $\color{#35bf28}+1.42\%$
test_serial 2.2117s 2.2103s 0.4524 Ops/s 0.4457 Ops/s $\color{#35bf28}+1.50\%$
test_parallel 1.9189s 1.8633s 0.5367 Ops/s 0.5328 Ops/s $\color{#35bf28}+0.73\%$
test_step_mdp_speed[True-True-True-True-True] 0.1753ms 42.0227μs 23.7967 KOps/s 22.9188 KOps/s $\color{#35bf28}+3.83\%$
test_step_mdp_speed[True-True-True-True-False] 51.2210μs 24.0234μs 41.6261 KOps/s 41.2884 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-True-True-False-True] 51.8110μs 23.7425μs 42.1186 KOps/s 41.7402 KOps/s $\color{#35bf28}+0.91\%$
test_step_mdp_speed[True-True-True-False-False] 39.2110μs 13.3655μs 74.8194 KOps/s 75.0477 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-True-False-True-True] 76.5120μs 44.7645μs 22.3391 KOps/s 21.6809 KOps/s $\color{#35bf28}+3.04\%$
test_step_mdp_speed[True-True-False-True-False] 53.0510μs 26.7446μs 37.3908 KOps/s 37.3226 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[True-True-False-False-True] 52.0410μs 26.1772μs 38.2012 KOps/s 37.5473 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-True-False-False-False] 44.1510μs 16.3623μs 61.1162 KOps/s 62.6222 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[True-False-True-True-True] 76.7310μs 47.0052μs 21.2742 KOps/s 20.5024 KOps/s $\color{#35bf28}+3.76\%$
test_step_mdp_speed[True-False-True-True-False] 59.3210μs 29.1485μs 34.3071 KOps/s 34.0415 KOps/s $\color{#35bf28}+0.78\%$
test_step_mdp_speed[True-False-True-False-True] 75.4020μs 27.0580μs 36.9577 KOps/s 37.2533 KOps/s $\color{#d91a1a}-0.79\%$
test_step_mdp_speed[True-False-True-False-False] 39.5910μs 15.8733μs 62.9987 KOps/s 62.5552 KOps/s $\color{#35bf28}+0.71\%$
test_step_mdp_speed[True-False-False-True-True] 87.7920μs 49.3211μs 20.2753 KOps/s 19.5495 KOps/s $\color{#35bf28}+3.71\%$
test_step_mdp_speed[True-False-False-True-False] 67.6510μs 31.2309μs 32.0196 KOps/s 31.4706 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-False-False-False-True] 57.2110μs 28.5745μs 34.9962 KOps/s 34.1533 KOps/s $\color{#35bf28}+2.47\%$
test_step_mdp_speed[True-False-False-False-False] 48.6100μs 18.6927μs 53.4969 KOps/s 54.2811 KOps/s $\color{#d91a1a}-1.44\%$
test_step_mdp_speed[False-True-True-True-True] 79.6120μs 46.9449μs 21.3015 KOps/s 20.7576 KOps/s $\color{#35bf28}+2.62\%$
test_step_mdp_speed[False-True-True-True-False] 57.7610μs 28.9997μs 34.4832 KOps/s 33.7692 KOps/s $\color{#35bf28}+2.11\%$
test_step_mdp_speed[False-True-True-False-True] 59.9610μs 30.0262μs 33.3043 KOps/s 33.3452 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-True-True-False-False] 44.4610μs 17.7172μs 56.4422 KOps/s 55.4601 KOps/s $\color{#35bf28}+1.77\%$
test_step_mdp_speed[False-True-False-True-True] 2.9039ms 50.2253μs 19.9103 KOps/s 19.4598 KOps/s $\color{#35bf28}+2.31\%$
test_step_mdp_speed[False-True-False-True-False] 64.0710μs 31.2761μs 31.9733 KOps/s 31.6960 KOps/s $\color{#35bf28}+0.87\%$
test_step_mdp_speed[False-True-False-False-True] 66.1120μs 32.2550μs 31.0029 KOps/s 30.8422 KOps/s $\color{#35bf28}+0.52\%$
test_step_mdp_speed[False-True-False-False-False] 53.0010μs 20.2803μs 49.3091 KOps/s 48.1970 KOps/s $\color{#35bf28}+2.31\%$
test_step_mdp_speed[False-False-True-True-True] 0.1097ms 51.0720μs 19.5802 KOps/s 18.5245 KOps/s $\textbf{\color{#35bf28}+5.70\%}$
test_step_mdp_speed[False-False-True-True-False] 64.1910μs 34.1734μs 29.2625 KOps/s 29.1989 KOps/s $\color{#35bf28}+0.22\%$
test_step_mdp_speed[False-False-True-False-True] 61.8410μs 32.5764μs 30.6970 KOps/s 29.9762 KOps/s $\color{#35bf28}+2.40\%$
test_step_mdp_speed[False-False-True-False-False] 51.4710μs 20.3422μs 49.1588 KOps/s 48.1234 KOps/s $\color{#35bf28}+2.15\%$
test_step_mdp_speed[False-False-False-True-True] 88.1410μs 54.1363μs 18.4719 KOps/s 17.9345 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[False-False-False-True-False] 78.4320μs 36.0550μs 27.7354 KOps/s 27.0455 KOps/s $\color{#35bf28}+2.55\%$
test_step_mdp_speed[False-False-False-False-True] 63.1010μs 34.9556μs 28.6077 KOps/s 28.7079 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[False-False-False-False-False] 50.0610μs 22.9183μs 43.6332 KOps/s 43.1625 KOps/s $\color{#35bf28}+1.09\%$
test_values[generalized_advantage_estimate-True-True] 21.0329ms 20.6318ms 48.4688 Ops/s 47.9376 Ops/s $\color{#35bf28}+1.11\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1443s 3.7874ms 264.0324 Ops/s 267.6601 Ops/s $\color{#d91a1a}-1.36\%$
test_values[td0_return_estimate-False-False] 0.1156ms 79.9261μs 12.5116 KOps/s 12.5165 KOps/s $\color{#d91a1a}-0.04\%$
test_values[td1_return_estimate-False-False] 51.0476ms 48.7507ms 20.5125 Ops/s 20.2297 Ops/s $\color{#35bf28}+1.40\%$
test_values[vec_td1_return_estimate-False-False] 1.3036ms 1.0822ms 924.0245 Ops/s 921.6794 Ops/s $\color{#35bf28}+0.25\%$
test_values[td_lambda_return_estimate-True-False] 83.4794ms 79.5296ms 12.5739 Ops/s 12.4997 Ops/s $\color{#35bf28}+0.59\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3514ms 1.0782ms 927.4756 Ops/s 926.7042 Ops/s $\color{#35bf28}+0.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 22.2370ms 21.9941ms 45.4667 Ops/s 47.5194 Ops/s $\color{#d91a1a}-4.32\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0400ms 0.7386ms 1.3539 KOps/s 1.3490 KOps/s $\color{#35bf28}+0.36\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7224ms 0.6763ms 1.4787 KOps/s 1.5205 KOps/s $\color{#d91a1a}-2.75\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5214ms 1.4715ms 679.5736 Ops/s 678.0166 Ops/s $\color{#35bf28}+0.23\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7557ms 0.6940ms 1.4409 KOps/s 1.4820 KOps/s $\color{#d91a1a}-2.77\%$
test_dqn_speed[False-None] 7.1390ms 1.5434ms 647.9071 Ops/s 660.0641 Ops/s $\color{#d91a1a}-1.84\%$
test_dqn_speed[False-backward] 2.2251ms 2.1474ms 465.6786 Ops/s 464.1875 Ops/s $\color{#35bf28}+0.32\%$
test_dqn_speed[True-None] 17.4325ms 0.5916ms 1.6905 KOps/s 1.7522 KOps/s $\color{#d91a1a}-3.52\%$
test_dqn_speed[True-backward] 1.2152ms 1.1580ms 863.5628 Ops/s 831.8052 Ops/s $\color{#35bf28}+3.82\%$
test_dqn_speed[reduce-overhead-None] 0.6583ms 0.6023ms 1.6604 KOps/s 1.6553 KOps/s $\color{#35bf28}+0.31\%$
test_dqn_speed[reduce-overhead-backward] 1.0518ms 0.9800ms 1.0204 KOps/s 1.0098 KOps/s $\color{#35bf28}+1.05\%$
test_ddpg_speed[False-None] 3.1387ms 2.8414ms 351.9411 Ops/s 351.3689 Ops/s $\color{#35bf28}+0.16\%$
test_ddpg_speed[False-backward] 4.5342ms 4.1497ms 240.9809 Ops/s 237.0160 Ops/s $\color{#35bf28}+1.67\%$
test_ddpg_speed[True-None] 1.4036ms 1.3545ms 738.2991 Ops/s 723.8572 Ops/s $\color{#35bf28}+2.00\%$
test_ddpg_speed[True-backward] 2.6098ms 2.5602ms 390.5876 Ops/s 382.6924 Ops/s $\color{#35bf28}+2.06\%$
test_ddpg_speed[reduce-overhead-None] 1.5360ms 1.3762ms 726.6213 Ops/s 720.1692 Ops/s $\color{#35bf28}+0.90\%$
test_ddpg_speed[reduce-overhead-backward] 0.2120s 0.1911s 5.2331 Ops/s 4.4367 Ops/s $\textbf{\color{#35bf28}+17.95\%}$
test_sac_speed[False-None] 8.3779ms 7.9341ms 126.0385 Ops/s 124.9110 Ops/s $\color{#35bf28}+0.90\%$
test_sac_speed[False-backward] 11.5157ms 10.9281ms 91.5068 Ops/s 89.9114 Ops/s $\color{#35bf28}+1.77\%$
test_sac_speed[True-None] 2.0604ms 1.8759ms 533.0816 Ops/s 525.9235 Ops/s $\color{#35bf28}+1.36\%$
test_sac_speed[True-backward] 3.7403ms 3.7017ms 270.1428 Ops/s 253.4993 Ops/s $\textbf{\color{#35bf28}+6.57\%}$
test_sac_speed[reduce-overhead-None] 19.9698ms 11.3178ms 88.3561 Ops/s 87.6542 Ops/s $\color{#35bf28}+0.80\%$
test_sac_speed[reduce-overhead-backward] 1.6555ms 1.6154ms 619.0523 Ops/s 619.5648 Ops/s $\color{#d91a1a}-0.08\%$
test_redq_deprec_speed[False-None] 9.2796ms 8.8903ms 112.4820 Ops/s 110.7139 Ops/s $\color{#35bf28}+1.60\%$
test_redq_deprec_speed[False-backward] 12.3417ms 12.1069ms 82.5973 Ops/s 81.4281 Ops/s $\color{#35bf28}+1.44\%$
test_redq_deprec_speed[True-None] 2.6660ms 2.5066ms 398.9473 Ops/s 394.3416 Ops/s $\color{#35bf28}+1.17\%$
test_redq_deprec_speed[True-backward] 4.5339ms 4.3979ms 227.3823 Ops/s 227.4639 Ops/s $\color{#d91a1a}-0.04\%$
test_redq_deprec_speed[reduce-overhead-None] 2.6922ms 2.5144ms 397.7131 Ops/s 392.2931 Ops/s $\color{#35bf28}+1.38\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.4643ms 4.3569ms 229.5212 Ops/s 226.1162 Ops/s $\color{#35bf28}+1.51\%$
test_td3_speed[False-None] 8.0692ms 7.8583ms 127.2540 Ops/s 126.4190 Ops/s $\color{#35bf28}+0.66\%$
test_td3_speed[False-backward] 11.0337ms 10.3883ms 96.2625 Ops/s 95.8249 Ops/s $\color{#35bf28}+0.46\%$
test_td3_speed[True-None] 1.7264ms 1.7004ms 588.0919 Ops/s 573.1646 Ops/s $\color{#35bf28}+2.60\%$
test_td3_speed[True-backward] 3.7739ms 3.3502ms 298.4910 Ops/s 293.6653 Ops/s $\color{#35bf28}+1.64\%$
test_td3_speed[reduce-overhead-None] 48.4549ms 24.8926ms 40.1725 Ops/s 40.0013 Ops/s $\color{#35bf28}+0.43\%$
test_td3_speed[reduce-overhead-backward] 1.3209ms 1.2685ms 788.3232 Ops/s 768.9744 Ops/s $\color{#35bf28}+2.52\%$
test_cql_speed[False-None] 17.0091ms 16.5981ms 60.2478 Ops/s 59.9293 Ops/s $\color{#35bf28}+0.53\%$
test_cql_speed[False-backward] 22.3163ms 21.8654ms 45.7343 Ops/s 45.0009 Ops/s $\color{#35bf28}+1.63\%$
test_cql_speed[True-None] 3.4698ms 3.3786ms 295.9798 Ops/s 284.2799 Ops/s $\color{#35bf28}+4.12\%$
test_cql_speed[True-backward] 5.7622ms 5.7255ms 174.6563 Ops/s 165.7495 Ops/s $\textbf{\color{#35bf28}+5.37\%}$
test_cql_speed[reduce-overhead-None] 20.2947ms 12.4554ms 80.2864 Ops/s 79.7079 Ops/s $\color{#35bf28}+0.73\%$
test_cql_speed[reduce-overhead-backward] 1.9925ms 1.7164ms 582.6106 Ops/s 553.4634 Ops/s $\textbf{\color{#35bf28}+5.27\%}$
test_a2c_speed[False-None] 3.2302ms 3.1477ms 317.6901 Ops/s 316.8156 Ops/s $\color{#35bf28}+0.28\%$
test_a2c_speed[False-backward] 6.7997ms 6.1618ms 162.2912 Ops/s 158.0008 Ops/s $\color{#35bf28}+2.72\%$
test_a2c_speed[True-None] 1.3730ms 1.3166ms 759.5542 Ops/s 761.1575 Ops/s $\color{#d91a1a}-0.21\%$
test_a2c_speed[True-backward] 3.0895ms 3.0220ms 330.9025 Ops/s 309.3285 Ops/s $\textbf{\color{#35bf28}+6.97\%}$
test_a2c_speed[reduce-overhead-None] 15.4655ms 8.7342ms 114.4919 Ops/s 116.1915 Ops/s $\color{#d91a1a}-1.46\%$
test_a2c_speed[reduce-overhead-backward] 1.4590ms 1.3933ms 717.7247 Ops/s 642.5487 Ops/s $\textbf{\color{#35bf28}+11.70\%}$
test_ppo_speed[False-None] 3.9185ms 3.7422ms 267.2232 Ops/s 263.3171 Ops/s $\color{#35bf28}+1.48\%$
test_ppo_speed[False-backward] 7.3088ms 6.8494ms 145.9977 Ops/s 141.1136 Ops/s $\color{#35bf28}+3.46\%$
test_ppo_speed[True-None] 1.5099ms 1.4171ms 705.6733 Ops/s 685.0318 Ops/s $\color{#35bf28}+3.01\%$
test_ppo_speed[True-backward] 3.3378ms 3.2277ms 309.8138 Ops/s 305.9150 Ops/s $\color{#35bf28}+1.27\%$
test_ppo_speed[reduce-overhead-None] 1.4561ms 1.4029ms 712.7997 Ops/s 697.4434 Ops/s $\color{#35bf28}+2.20\%$
test_ppo_speed[reduce-overhead-backward] 3.2486ms 3.1835ms 314.1187 Ops/s 308.1378 Ops/s $\color{#35bf28}+1.94\%$
test_reinforce_speed[False-None] 2.3420ms 2.2523ms 443.9915 Ops/s 439.5409 Ops/s $\color{#35bf28}+1.01\%$
test_reinforce_speed[False-backward] 3.6951ms 3.2802ms 304.8589 Ops/s 298.2915 Ops/s $\color{#35bf28}+2.20\%$
test_reinforce_speed[True-None] 1.3514ms 1.2690ms 788.0213 Ops/s 767.4264 Ops/s $\color{#35bf28}+2.68\%$
test_reinforce_speed[True-backward] 3.1200ms 3.0401ms 328.9363 Ops/s 320.6591 Ops/s $\color{#35bf28}+2.58\%$
test_reinforce_speed[reduce-overhead-None] 18.9339ms 10.3993ms 96.1601 Ops/s 98.5504 Ops/s $\color{#d91a1a}-2.43\%$
test_reinforce_speed[reduce-overhead-backward] 1.5007ms 1.4471ms 691.0323 Ops/s 675.0287 Ops/s $\color{#35bf28}+2.37\%$
test_iql_speed[False-None] 9.9964ms 9.2350ms 108.2832 Ops/s 107.9111 Ops/s $\color{#35bf28}+0.34\%$
test_iql_speed[False-backward] 13.5205ms 13.0442ms 76.6622 Ops/s 76.0641 Ops/s $\color{#35bf28}+0.79\%$
test_iql_speed[True-None] 2.3445ms 2.2669ms 441.1234 Ops/s 434.3629 Ops/s $\color{#35bf28}+1.56\%$
test_iql_speed[True-backward] 5.3842ms 4.9297ms 202.8507 Ops/s 196.6323 Ops/s $\color{#35bf28}+3.16\%$
test_iql_speed[reduce-overhead-None] 19.0062ms 10.6580ms 93.8267 Ops/s 93.2058 Ops/s $\color{#35bf28}+0.67\%$
test_iql_speed[reduce-overhead-backward] 1.8655ms 1.8114ms 552.0636 Ops/s 529.8135 Ops/s $\color{#35bf28}+4.20\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4556ms 5.9898ms 166.9519 Ops/s 163.9596 Ops/s $\color{#35bf28}+1.82\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6027ms 0.3503ms 2.8544 KOps/s 3.4013 KOps/s $\textbf{\color{#d91a1a}-16.08\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4831ms 0.2435ms 4.1066 KOps/s 3.7562 KOps/s $\textbf{\color{#35bf28}+9.33\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9832ms 5.7678ms 173.3759 Ops/s 171.6539 Ops/s $\color{#35bf28}+1.00\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6958ms 0.2758ms 3.6255 KOps/s 3.8541 KOps/s $\textbf{\color{#d91a1a}-5.93\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5204ms 0.2475ms 4.0408 KOps/s 3.4251 KOps/s $\textbf{\color{#35bf28}+17.98\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4954ms 1.2738ms 785.0765 Ops/s 786.5967 Ops/s $\color{#d91a1a}-0.19\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5827ms 1.2926ms 773.6308 Ops/s 802.6170 Ops/s $\color{#d91a1a}-3.61\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0390ms 5.8854ms 169.9133 Ops/s 166.2438 Ops/s $\color{#35bf28}+2.21\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1581ms 0.3961ms 2.5243 KOps/s 2.3299 KOps/s $\textbf{\color{#35bf28}+8.34\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6634ms 0.4058ms 2.4641 KOps/s 2.0385 KOps/s $\textbf{\color{#35bf28}+20.88\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8390ms 5.7286ms 174.5623 Ops/s 172.6870 Ops/s $\color{#35bf28}+1.09\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8220ms 0.3133ms 3.1917 KOps/s 3.0365 KOps/s $\textbf{\color{#35bf28}+5.11\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5500ms 0.3227ms 3.0992 KOps/s 4.0091 KOps/s $\textbf{\color{#d91a1a}-22.70\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9075ms 5.6678ms 176.4348 Ops/s 173.0734 Ops/s $\color{#35bf28}+1.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8198ms 0.3100ms 3.2259 KOps/s 2.9866 KOps/s $\textbf{\color{#35bf28}+8.01\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5268ms 0.2962ms 3.3756 KOps/s 4.0795 KOps/s $\textbf{\color{#d91a1a}-17.25\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.9845ms 5.8213ms 171.7818 Ops/s 169.4770 Ops/s $\color{#35bf28}+1.36\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7673ms 0.4600ms 2.1741 KOps/s 2.2133 KOps/s $\color{#d91a1a}-1.77\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6696ms 0.4262ms 2.3465 KOps/s 2.2584 KOps/s $\color{#35bf28}+3.90\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.9516ms 5.3808ms 185.8446 Ops/s 49.7565 Ops/s $\textbf{\color{#35bf28}+273.51\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.1912ms 2.0945ms 477.4499 Ops/s 460.0320 Ops/s $\color{#35bf28}+3.79\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.5547ms 1.2965ms 771.3211 Ops/s 760.2932 Ops/s $\color{#35bf28}+1.45\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.3635ms 5.5163ms 181.2821 Ops/s 179.4036 Ops/s $\color{#35bf28}+1.05\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.9706ms 2.1026ms 475.5953 Ops/s 461.0365 Ops/s $\color{#35bf28}+3.16\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.9618ms 0.9816ms 1.0187 KOps/s 973.3184 Ops/s $\color{#35bf28}+4.67\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5224s 16.0088ms 62.4657 Ops/s 170.3891 Ops/s $\textbf{\color{#d91a1a}-63.34\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 11.2999ms 2.2874ms 437.1758 Ops/s 62.9604 Ops/s $\textbf{\color{#35bf28}+594.37\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.2356ms 1.2918ms 774.1166 Ops/s 783.5672 Ops/s $\color{#d91a1a}-1.21\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 61.7881ms 58.0279ms 17.2331 Ops/s 17.3392 Ops/s $\color{#d91a1a}-0.61\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 17.8502ms 16.4029ms 60.9647 Ops/s 60.2429 Ops/s $\color{#35bf28}+1.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 61.8502ms 57.2699ms 17.4612 Ops/s 16.8884 Ops/s $\color{#35bf28}+3.39\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 17.8187ms 16.4469ms 60.8018 Ops/s 59.6344 Ops/s $\color{#35bf28}+1.96\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 61.2076ms 57.4970ms 17.3922 Ops/s 17.2805 Ops/s $\color{#35bf28}+0.65\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.1589ms 17.8158ms 56.1298 Ops/s 55.8616 Ops/s $\color{#35bf28}+0.48\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

BE Better errors, logs, docs or test utils 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

点击 这是indexloc提供的php浏览器服务,不要输入任何密码和下载