Skip to content

[CI] Improve RST section underline checker#3539

Merged
vmoens merged 2 commits intomainfrom
ci/improve-rst-underline-check
Mar 2, 2026
Merged

[CI] Improve RST section underline checker#3539
vmoens merged 2 commits intomainfrom
ci/improve-rst-underline-check

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Mar 2, 2026

Rewrite the check-sphinx-section-underline script with a line-by-line
scanner instead of a regex approach. Handles overline+title+underline
patterns, skips doctest blocks (>>> / ...), and strips BOM characters
from generated RST files to avoid false positives.

Also adds --fix support via argparse.

Made-with: Cursor

Rewrite the check-sphinx-section-underline script with a line-by-line
scanner instead of a regex approach. Handles overline+title+underline
patterns, skips doctest blocks (>>> / ...), and strips BOM characters
from generated RST files to avoid false positives.

Also adds --fix support via argparse.

Made-with: Cursor
@pytorch-bot
Copy link

pytorch-bot bot commented Mar 2, 2026

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3539

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.

@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 Mar 2, 2026
@github-actions github-actions bot added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Mar 2, 2026
@github-actions github-actions bot added the Benchmarks rl/benchmark changes label Mar 2, 2026
@vmoens vmoens merged commit ff77695 into main Mar 2, 2026
103 of 104 checks passed
@vmoens vmoens deleted the ci/improve-rst-underline-check branch March 2, 2026 11:34
@github-actions
Copy link
Contributor

github-actions bot commented Mar 2, 2026

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

Total Benchmarks: 173. Improved: $\large\color{#35bf28}17$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 80.4049μs 79.2054μs 12.6254 KOps/s 12.4610 KOps/s $\color{#35bf28}+1.32\%$
test_tensor_to_bytestream_speed[torch.save] 0.1397ms 0.1391ms 7.1913 KOps/s 7.1890 KOps/s $\color{#35bf28}+0.03\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1326s 0.1324s 7.5524 Ops/s 7.4518 Ops/s $\color{#35bf28}+1.35\%$
test_tensor_to_bytestream_speed[numpy] 2.5746μs 2.5656μs 389.7737 KOps/s 396.6767 KOps/s $\color{#d91a1a}-1.74\%$
test_tensor_to_bytestream_speed[safetensors] 40.3984μs 39.6256μs 25.2362 KOps/s 26.8056 KOps/s $\textbf{\color{#d91a1a}-5.85\%}$
test_simple 0.5547s 0.5536s 1.8064 Ops/s 1.7388 Ops/s $\color{#35bf28}+3.89\%$
test_transformed 1.1051s 1.1034s 0.9063 Ops/s 0.8840 Ops/s $\color{#35bf28}+2.52\%$
test_serial 1.7206s 1.7191s 0.5817 Ops/s 0.5778 Ops/s $\color{#35bf28}+0.67\%$
test_parallel 1.1587s 1.0715s 0.9333 Ops/s 0.9556 Ops/s $\color{#d91a1a}-2.33\%$
test_step_mdp_speed[True-True-True-True-True] 0.2571ms 42.7924μs 23.3686 KOps/s 23.3351 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[True-True-True-True-False] 70.2010μs 24.2402μs 41.2539 KOps/s 41.8811 KOps/s $\color{#d91a1a}-1.50\%$
test_step_mdp_speed[True-True-True-False-True] 53.3810μs 23.6870μs 42.2172 KOps/s 41.3915 KOps/s $\color{#35bf28}+2.00\%$
test_step_mdp_speed[True-True-True-False-False] 42.6600μs 13.2135μs 75.6802 KOps/s 75.9154 KOps/s $\color{#d91a1a}-0.31\%$
test_step_mdp_speed[True-True-False-True-True] 82.1910μs 45.6319μs 21.9145 KOps/s 21.3533 KOps/s $\color{#35bf28}+2.63\%$
test_step_mdp_speed[True-True-False-True-False] 56.9610μs 26.6804μs 37.4808 KOps/s 38.2486 KOps/s $\color{#d91a1a}-2.01\%$
test_step_mdp_speed[True-True-False-False-True] 61.7410μs 26.5112μs 37.7199 KOps/s 37.2461 KOps/s $\color{#35bf28}+1.27\%$
test_step_mdp_speed[True-True-False-False-False] 49.0300μs 15.9153μs 62.8325 KOps/s 63.1903 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[True-False-True-True-True] 80.3710μs 49.6778μs 20.1297 KOps/s 20.3975 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[True-False-True-True-False] 67.2300μs 29.8175μs 33.5374 KOps/s 34.0072 KOps/s $\color{#d91a1a}-1.38\%$
test_step_mdp_speed[True-False-True-False-True] 62.4800μs 26.6787μs 37.4830 KOps/s 37.3881 KOps/s $\color{#35bf28}+0.25\%$
test_step_mdp_speed[True-False-True-False-False] 48.2000μs 16.1201μs 62.0345 KOps/s 62.7251 KOps/s $\color{#d91a1a}-1.10\%$
test_step_mdp_speed[True-False-False-True-True] 0.1009ms 50.5855μs 19.7685 KOps/s 19.5559 KOps/s $\color{#35bf28}+1.09\%$
test_step_mdp_speed[True-False-False-True-False] 62.8310μs 31.7264μs 31.5195 KOps/s 31.6479 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[True-False-False-False-True] 64.6900μs 29.6399μs 33.7383 KOps/s 34.1882 KOps/s $\color{#d91a1a}-1.32\%$
test_step_mdp_speed[True-False-False-False-False] 61.6900μs 18.8518μs 53.0452 KOps/s 53.3633 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[False-True-True-True-True] 79.5610μs 49.0462μs 20.3889 KOps/s 20.0551 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[False-True-True-True-False] 63.4710μs 29.5071μs 33.8901 KOps/s 33.4592 KOps/s $\color{#35bf28}+1.29\%$
test_step_mdp_speed[False-True-True-False-True] 2.4661ms 32.0246μs 31.2260 KOps/s 32.7398 KOps/s $\color{#d91a1a}-4.62\%$
test_step_mdp_speed[False-True-True-False-False] 48.2100μs 17.8295μs 56.0867 KOps/s 57.0056 KOps/s $\color{#d91a1a}-1.61\%$
test_step_mdp_speed[False-True-False-True-True] 82.8510μs 50.3152μs 19.8747 KOps/s 19.5446 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[False-True-False-True-False] 63.6810μs 31.7921μs 31.4543 KOps/s 31.1756 KOps/s $\color{#35bf28}+0.89\%$
test_step_mdp_speed[False-True-False-False-True] 56.9300μs 32.7714μs 30.5144 KOps/s 29.6621 KOps/s $\color{#35bf28}+2.87\%$
test_step_mdp_speed[False-True-False-False-False] 53.4110μs 20.2511μs 49.3800 KOps/s 50.0440 KOps/s $\color{#d91a1a}-1.33\%$
test_step_mdp_speed[False-False-True-True-True] 87.1700μs 53.7582μs 18.6018 KOps/s 18.4885 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[False-False-True-True-False] 65.6000μs 34.8059μs 28.7308 KOps/s 28.6325 KOps/s $\color{#35bf28}+0.34\%$
test_step_mdp_speed[False-False-True-False-True] 70.4700μs 33.8143μs 29.5733 KOps/s 30.0610 KOps/s $\color{#d91a1a}-1.62\%$
test_step_mdp_speed[False-False-True-False-False] 48.9000μs 20.0867μs 49.7842 KOps/s 49.2927 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[False-False-False-True-True] 91.6510μs 55.2446μs 18.1013 KOps/s 17.7963 KOps/s $\color{#35bf28}+1.71\%$
test_step_mdp_speed[False-False-False-True-False] 70.2700μs 37.2326μs 26.8582 KOps/s 26.8637 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[False-False-False-False-True] 72.5110μs 35.8619μs 27.8847 KOps/s 28.8025 KOps/s $\color{#d91a1a}-3.19\%$
test_step_mdp_speed[False-False-False-False-False] 52.6710μs 22.7695μs 43.9184 KOps/s 44.3964 KOps/s $\color{#d91a1a}-1.08\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8485s 0.7536s 1.3269 Ops/s 1.3380 Ops/s $\color{#d91a1a}-0.83\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7122s 0.6160s 1.6235 Ops/s 1.6224 Ops/s $\color{#35bf28}+0.07\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7433s 1.6647s 0.6007 Ops/s 0.6068 Ops/s $\color{#d91a1a}-1.00\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5154s 1.4379s 0.6955 Ops/s 0.7019 Ops/s $\color{#d91a1a}-0.92\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0072s 1.9259s 0.5192 Ops/s 0.5251 Ops/s $\color{#d91a1a}-1.11\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7822s 1.7031s 0.5872 Ops/s 0.5942 Ops/s $\color{#d91a1a}-1.19\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8209s 4.6921s 0.2131 Ops/s 0.2142 Ops/s $\color{#d91a1a}-0.50\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4864s 4.4202s 0.2262 Ops/s 0.2249 Ops/s $\color{#35bf28}+0.61\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9561s 1.8823s 0.5313 Ops/s 0.5317 Ops/s $\color{#d91a1a}-0.09\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7482s 1.6247s 0.6155 Ops/s 0.6205 Ops/s $\color{#d91a1a}-0.80\%$
test_values[generalized_advantage_estimate-True-True] 10.8121ms 10.6903ms 93.5427 Ops/s 94.9230 Ops/s $\color{#d91a1a}-1.45\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.3890ms 17.6572ms 56.6342 Ops/s 89.7915 Ops/s $\textbf{\color{#d91a1a}-36.93\%}$
test_values[td0_return_estimate-False-False] 0.2167ms 0.1315ms 7.6035 KOps/s 7.8282 KOps/s $\color{#d91a1a}-2.87\%$
test_values[td1_return_estimate-False-False] 31.5098ms 30.2342ms 33.0752 Ops/s 34.8528 Ops/s $\textbf{\color{#d91a1a}-5.10\%}$
test_values[vec_td1_return_estimate-False-False] 18.8644ms 17.7715ms 56.2698 Ops/s 88.9009 Ops/s $\textbf{\color{#d91a1a}-36.71\%}$
test_values[td_lambda_return_estimate-True-False] 46.6129ms 44.7070ms 22.3679 Ops/s 23.4148 Ops/s $\color{#d91a1a}-4.47\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.8953ms 17.7749ms 56.2590 Ops/s 88.4438 Ops/s $\textbf{\color{#d91a1a}-36.39\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.9961ms 9.5587ms 104.6162 Ops/s 107.6342 Ops/s $\color{#d91a1a}-2.80\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8767ms 1.5627ms 639.9311 Ops/s 639.3835 Ops/s $\color{#35bf28}+0.09\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5031ms 0.4402ms 2.2719 KOps/s 2.3446 KOps/s $\color{#d91a1a}-3.10\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.2991ms 34.6408ms 28.8677 Ops/s 54.1861 Ops/s $\textbf{\color{#d91a1a}-46.72\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9130ms 1.7551ms 569.7558 Ops/s 565.6542 Ops/s $\color{#35bf28}+0.73\%$
test_dqn_speed[False-None] 1.5685ms 1.4160ms 706.1966 Ops/s 699.7129 Ops/s $\color{#35bf28}+0.93\%$
test_dqn_speed[False-backward] 1.9908ms 1.9340ms 517.0612 Ops/s 506.6501 Ops/s $\color{#35bf28}+2.05\%$
test_dqn_speed[True-None] 0.9242ms 0.5640ms 1.7731 KOps/s 1.7700 KOps/s $\color{#35bf28}+0.18\%$
test_dqn_speed[True-backward] 1.1234ms 1.0213ms 979.1096 Ops/s 896.6134 Ops/s $\textbf{\color{#35bf28}+9.20\%}$
test_dqn_speed[reduce-overhead-None] 0.6725ms 0.5414ms 1.8471 KOps/s 1.7626 KOps/s $\color{#35bf28}+4.80\%$
test_ddpg_speed[False-None] 3.3441ms 2.8834ms 346.8093 Ops/s 346.4179 Ops/s $\color{#35bf28}+0.11\%$
test_ddpg_speed[False-backward] 4.2588ms 4.0991ms 243.9571 Ops/s 244.1653 Ops/s $\color{#d91a1a}-0.09\%$
test_ddpg_speed[True-None] 1.8370ms 1.4252ms 701.6408 Ops/s 702.4727 Ops/s $\color{#d91a1a}-0.12\%$
test_ddpg_speed[True-backward] 2.4395ms 2.3986ms 416.9092 Ops/s 374.3502 Ops/s $\textbf{\color{#35bf28}+11.37\%}$
test_ddpg_speed[reduce-overhead-None] 1.5769ms 1.4152ms 706.6282 Ops/s 700.7421 Ops/s $\color{#35bf28}+0.84\%$
test_sac_speed[False-None] 8.7725ms 8.1686ms 122.4206 Ops/s 121.8061 Ops/s $\color{#35bf28}+0.50\%$
test_sac_speed[False-backward] 12.3686ms 11.5938ms 86.2527 Ops/s 88.2246 Ops/s $\color{#d91a1a}-2.24\%$
test_sac_speed[True-None] 2.3828ms 2.1648ms 461.9424 Ops/s 458.6025 Ops/s $\color{#35bf28}+0.73\%$
test_sac_speed[True-backward] 4.1602ms 3.9937ms 250.3923 Ops/s 214.9966 Ops/s $\textbf{\color{#35bf28}+16.46\%}$
test_sac_speed[reduce-overhead-None] 2.5087ms 2.1456ms 466.0631 Ops/s 454.8594 Ops/s $\color{#35bf28}+2.46\%$
test_redq_speed[False-None] 15.5734ms 10.9532ms 91.2978 Ops/s 95.9830 Ops/s $\color{#d91a1a}-4.88\%$
test_redq_speed[False-backward] 18.8099ms 17.9585ms 55.6839 Ops/s 56.6729 Ops/s $\color{#d91a1a}-1.74\%$
test_redq_speed[True-None] 4.5797ms 4.2902ms 233.0871 Ops/s 233.6823 Ops/s $\color{#d91a1a}-0.25\%$
test_redq_speed[True-backward] 9.9408ms 9.5993ms 104.1748 Ops/s 107.7122 Ops/s $\color{#d91a1a}-3.28\%$
test_redq_speed[reduce-overhead-None] 4.7425ms 4.2165ms 237.1608 Ops/s 222.3370 Ops/s $\textbf{\color{#35bf28}+6.67\%}$
test_redq_deprec_speed[False-None] 11.6723ms 11.1218ms 89.9138 Ops/s 91.1837 Ops/s $\color{#d91a1a}-1.39\%$
test_redq_deprec_speed[False-backward] 16.3321ms 15.9056ms 62.8710 Ops/s 63.9459 Ops/s $\color{#d91a1a}-1.68\%$
test_redq_deprec_speed[True-None] 3.7893ms 3.5743ms 279.7749 Ops/s 274.4212 Ops/s $\color{#35bf28}+1.95\%$
test_redq_deprec_speed[True-backward] 7.7488ms 7.4033ms 135.0757 Ops/s 128.6518 Ops/s $\color{#35bf28}+4.99\%$
test_redq_deprec_speed[reduce-overhead-None] 3.9745ms 3.5337ms 282.9880 Ops/s 264.6089 Ops/s $\textbf{\color{#35bf28}+6.95\%}$
test_td3_speed[False-None] 9.7370ms 8.2699ms 120.9203 Ops/s 121.8288 Ops/s $\color{#d91a1a}-0.75\%$
test_td3_speed[False-backward] 11.5169ms 11.0309ms 90.6547 Ops/s 89.4552 Ops/s $\color{#35bf28}+1.34\%$
test_td3_speed[True-None] 1.8797ms 1.8188ms 549.7992 Ops/s 548.9036 Ops/s $\color{#35bf28}+0.16\%$
test_td3_speed[True-backward] 3.6581ms 3.5655ms 280.4689 Ops/s 275.2307 Ops/s $\color{#35bf28}+1.90\%$
test_td3_speed[reduce-overhead-None] 1.8331ms 1.7770ms 562.7563 Ops/s 557.6131 Ops/s $\color{#35bf28}+0.92\%$
test_cql_speed[False-None] 28.9122ms 26.5916ms 37.6059 Ops/s 38.4386 Ops/s $\color{#d91a1a}-2.17\%$
test_cql_speed[False-backward] 36.2886ms 35.5330ms 28.1429 Ops/s 27.7730 Ops/s $\color{#35bf28}+1.33\%$
test_cql_speed[True-None] 12.6856ms 12.3029ms 81.2817 Ops/s 81.5165 Ops/s $\color{#d91a1a}-0.29\%$
test_cql_speed[True-backward] 18.8546ms 18.2493ms 54.7965 Ops/s 56.4729 Ops/s $\color{#d91a1a}-2.97\%$
test_cql_speed[reduce-overhead-None] 13.8230ms 12.3466ms 80.9938 Ops/s 81.8585 Ops/s $\color{#d91a1a}-1.06\%$
test_a2c_speed[False-None] 5.6766ms 5.4701ms 182.8136 Ops/s 182.0674 Ops/s $\color{#35bf28}+0.41\%$
test_a2c_speed[False-backward] 12.1796ms 11.8444ms 84.4280 Ops/s 85.5630 Ops/s $\color{#d91a1a}-1.33\%$
test_a2c_speed[True-None] 4.2895ms 3.7555ms 266.2789 Ops/s 264.5336 Ops/s $\color{#35bf28}+0.66\%$
test_a2c_speed[True-backward] 8.7046ms 8.4756ms 117.9859 Ops/s 106.5300 Ops/s $\textbf{\color{#35bf28}+10.75\%}$
test_a2c_speed[reduce-overhead-None] 3.9290ms 3.7360ms 267.6669 Ops/s 269.2710 Ops/s $\color{#d91a1a}-0.60\%$
test_ppo_speed[False-None] 6.2100ms 5.9764ms 167.3241 Ops/s 167.5913 Ops/s $\color{#d91a1a}-0.16\%$
test_ppo_speed[False-backward] 12.8515ms 12.5117ms 79.9253 Ops/s 80.6016 Ops/s $\color{#d91a1a}-0.84\%$
test_ppo_speed[True-None] 3.8186ms 3.6516ms 273.8522 Ops/s 268.9896 Ops/s $\color{#35bf28}+1.81\%$
test_ppo_speed[True-backward] 8.4990ms 8.3176ms 120.2275 Ops/s 109.5009 Ops/s $\textbf{\color{#35bf28}+9.80\%}$
test_ppo_speed[reduce-overhead-None] 3.8986ms 3.6213ms 276.1420 Ops/s 273.9796 Ops/s $\color{#35bf28}+0.79\%$
test_reinforce_speed[False-None] 5.0287ms 4.5537ms 219.6017 Ops/s 215.7608 Ops/s $\color{#35bf28}+1.78\%$
test_reinforce_speed[False-backward] 7.5812ms 7.3471ms 136.1088 Ops/s 136.6748 Ops/s $\color{#d91a1a}-0.41\%$
test_reinforce_speed[True-None] 3.1141ms 2.9050ms 344.2358 Ops/s 342.0972 Ops/s $\color{#35bf28}+0.63\%$
test_reinforce_speed[True-backward] 8.0937ms 7.6870ms 130.0892 Ops/s 130.9306 Ops/s $\color{#d91a1a}-0.64\%$
test_reinforce_speed[reduce-overhead-None] 3.0160ms 2.8666ms 348.8410 Ops/s 348.0615 Ops/s $\color{#35bf28}+0.22\%$
test_iql_speed[False-None] 24.1387ms 20.0561ms 49.8601 Ops/s 48.3525 Ops/s $\color{#35bf28}+3.12\%$
test_iql_speed[False-backward] 30.9870ms 30.4030ms 32.8914 Ops/s 32.6031 Ops/s $\color{#35bf28}+0.88\%$
test_iql_speed[True-None] 8.7245ms 8.4139ms 118.8506 Ops/s 116.3257 Ops/s $\color{#35bf28}+2.17\%$
test_iql_speed[True-backward] 17.1350ms 16.5011ms 60.6020 Ops/s 58.4578 Ops/s $\color{#35bf28}+3.67\%$
test_iql_speed[reduce-overhead-None] 8.6835ms 8.4273ms 118.6625 Ops/s 116.0287 Ops/s $\color{#35bf28}+2.27\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2418ms 6.0402ms 165.5579 Ops/s 166.0570 Ops/s $\color{#d91a1a}-0.30\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.9641ms 0.3486ms 2.8684 KOps/s 3.5002 KOps/s $\textbf{\color{#d91a1a}-18.05\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6506ms 0.3400ms 2.9410 KOps/s 3.7222 KOps/s $\textbf{\color{#d91a1a}-20.99\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2012ms 5.8890ms 169.8086 Ops/s 170.3283 Ops/s $\color{#d91a1a}-0.31\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7260ms 0.3201ms 3.1245 KOps/s 3.5932 KOps/s $\textbf{\color{#d91a1a}-13.05\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4827ms 0.2644ms 3.7818 KOps/s 3.7982 KOps/s $\color{#d91a1a}-0.43\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6777ms 1.4078ms 710.3414 Ops/s 772.1629 Ops/s $\textbf{\color{#d91a1a}-8.01\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.7654ms 1.3273ms 753.4117 Ops/s 828.9987 Ops/s $\textbf{\color{#d91a1a}-9.12\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.1176ms 6.2096ms 161.0416 Ops/s 164.8821 Ops/s $\color{#d91a1a}-2.33\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8916ms 0.4675ms 2.1388 KOps/s 1.8676 KOps/s $\textbf{\color{#35bf28}+14.53\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8833ms 0.4179ms 2.3931 KOps/s 2.1015 KOps/s $\textbf{\color{#35bf28}+13.88\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.3490ms 5.9045ms 169.3609 Ops/s 169.7345 Ops/s $\color{#d91a1a}-0.22\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0783ms 0.3272ms 3.0561 KOps/s 3.0146 KOps/s $\color{#35bf28}+1.38\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5301ms 0.3333ms 3.0001 KOps/s 3.7288 KOps/s $\textbf{\color{#d91a1a}-19.54\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.4008ms 5.8740ms 170.2410 Ops/s 172.1427 Ops/s $\color{#d91a1a}-1.10\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9003ms 0.2844ms 3.5156 KOps/s 3.5832 KOps/s $\color{#d91a1a}-1.89\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5265ms 0.3215ms 3.1107 KOps/s 3.8072 KOps/s $\textbf{\color{#d91a1a}-18.29\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5408ms 6.0531ms 165.2041 Ops/s 165.7466 Ops/s $\color{#d91a1a}-0.33\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7337ms 0.4377ms 2.2847 KOps/s 1.9887 KOps/s $\textbf{\color{#35bf28}+14.88\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7151ms 0.4615ms 2.1670 KOps/s 2.1003 KOps/s $\color{#35bf28}+3.17\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.6080s 17.0980ms 58.4864 Ops/s 52.8310 Ops/s $\textbf{\color{#35bf28}+10.70\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 12.2614ms 2.0661ms 484.0094 Ops/s 530.5697 Ops/s $\textbf{\color{#d91a1a}-8.78\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.0678ms 0.9012ms 1.1096 KOps/s 760.2130 Ops/s $\textbf{\color{#35bf28}+45.96\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 10.2021ms 5.0685ms 197.2970 Ops/s 196.8340 Ops/s $\color{#35bf28}+0.24\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 4.0645ms 1.7926ms 557.8626 Ops/s 528.7981 Ops/s $\textbf{\color{#35bf28}+5.50\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.0605ms 0.9187ms 1.0885 KOps/s 781.4830 Ops/s $\textbf{\color{#35bf28}+39.29\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5227s 15.6257ms 63.9972 Ops/s 59.1556 Ops/s $\textbf{\color{#35bf28}+8.18\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.2551ms 1.9531ms 512.0140 Ops/s 482.9610 Ops/s $\textbf{\color{#35bf28}+6.02\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0807ms 1.1148ms 897.0172 Ops/s 916.7403 Ops/s $\color{#d91a1a}-2.15\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 39.5874ms 36.5134ms 27.3872 Ops/s 27.3722 Ops/s $\color{#35bf28}+0.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 21.1446ms 18.8785ms 52.9703 Ops/s 54.2617 Ops/s $\color{#d91a1a}-2.38\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 42.0685ms 37.3573ms 26.7685 Ops/s 26.6876 Ops/s $\color{#35bf28}+0.30\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 21.0196ms 19.0203ms 52.5755 Ops/s 53.7165 Ops/s $\color{#d91a1a}-2.12\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 40.7866ms 39.0070ms 25.6364 Ops/s 25.2982 Ops/s $\color{#35bf28}+1.34\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 22.1311ms 20.3614ms 49.1126 Ops/s 49.6170 Ops/s $\color{#d91a1a}-1.02\%$
test_storage_write_lazystack[50-img_shape0-small] 0.9449ms 0.2193ms 4.5601 KOps/s 4.4374 KOps/s $\color{#35bf28}+2.77\%$
test_storage_write_lazystack[100-img_shape1-atari] 1.7026ms 1.3649ms 732.6702 Ops/s 719.5597 Ops/s $\color{#35bf28}+1.82\%$
test_storage_write_lazystack[100-img_shape2-large_img] 2.4663ms 2.3000ms 434.7913 Ops/s 418.8643 Ops/s $\color{#35bf28}+3.80\%$
test_storage_write_lazystack[200-img_shape3-large_batch] 3.2764ms 2.8569ms 350.0335 Ops/s 345.7176 Ops/s $\color{#35bf28}+1.25\%$
test_storage_write_contiguous[50-img_shape0-small] 0.5720ms 0.1391ms 7.1878 KOps/s 7.2108 KOps/s $\color{#d91a1a}-0.32\%$
test_storage_write_contiguous[100-img_shape1-atari] 0.3491ms 0.1887ms 5.2991 KOps/s 5.1210 KOps/s $\color{#35bf28}+3.48\%$
test_storage_write_contiguous[100-img_shape2-large_img] 1.9313ms 1.7662ms 566.1944 Ops/s 570.8738 Ops/s $\color{#d91a1a}-0.82\%$
test_storage_write_contiguous[200-img_shape3-large_batch] 1.4695ms 1.2958ms 771.7092 Ops/s 776.4734 Ops/s $\color{#d91a1a}-0.61\%$
test_collector_stack_then_write[50-img_shape0-small] 1.3403ms 1.1222ms 891.0868 Ops/s 890.6956 Ops/s $\color{#35bf28}+0.04\%$
test_collector_stack_then_write[100-img_shape1-atari] 7.5026ms 3.6082ms 277.1483 Ops/s 279.9574 Ops/s $\color{#d91a1a}-1.00\%$
test_collector_stack_then_write[100-img_shape2-large_img] 11.0974ms 5.7153ms 174.9700 Ops/s 176.2863 Ops/s $\color{#d91a1a}-0.75\%$
test_collector_stack_then_write[200-img_shape3-large_batch] 7.3767ms 6.8950ms 145.0322 Ops/s 143.9878 Ops/s $\color{#35bf28}+0.73\%$
test_collector_lazystack_then_write[50-img_shape0-small] 0.4445ms 0.2767ms 3.6134 KOps/s 3.5741 KOps/s $\color{#35bf28}+1.10\%$
test_collector_lazystack_then_write[100-img_shape1-atari] 1.9767ms 1.4974ms 667.8454 Ops/s 679.5456 Ops/s $\color{#d91a1a}-1.72\%$
test_collector_lazystack_then_write[100-img_shape2-large_img] 2.8506ms 2.4230ms 412.7181 Ops/s 391.5765 Ops/s $\textbf{\color{#35bf28}+5.40\%}$
test_collector_lazystack_then_write[200-img_shape3-large_batch] 3.5195ms 3.0640ms 326.3746 Ops/s 323.5870 Ops/s $\color{#35bf28}+0.86\%$
test_collector_without_rb[100-img_shape0-atari] 33.9584ms 33.3102ms 30.0208 Ops/s 30.0729 Ops/s $\color{#d91a1a}-0.17\%$
test_collector_without_rb[200-img_shape1-large_batch] 65.7155ms 65.2991ms 15.3141 Ops/s 15.3438 Ops/s $\color{#d91a1a}-0.19\%$
test_collector_with_rb[100-img_shape0-atari] 39.0690ms 38.5136ms 25.9649 Ops/s 26.2197 Ops/s $\color{#d91a1a}-0.97\%$
test_collector_with_rb[200-img_shape1-large_batch] 0.6831s 0.1190s 8.4035 Ops/s 13.4470 Ops/s $\textbf{\color{#d91a1a}-37.51\%}$

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

Labels

Benchmarks rl/benchmark changes CI Has to do with CI setup (e.g. wheels & builds, tests...) 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