Skip to content

[CI] Skip Windows-incompatible tests in optional deps CI#3348

Merged
vmoens merged 1 commit into
mainfrom
fix-windows-optdepts-ci
Jan 19, 2026
Merged

[CI] Skip Windows-incompatible tests in optional deps CI#3348
vmoens merged 1 commit into
mainfrom
fix-windows-optdepts-ci

Conversation

@vmoens

@vmoens vmoens commented Jan 19, 2026

Copy link
Copy Markdown
Collaborator

Summary

  • Skip TestValueFunctionCompile class on Windows (torch.compile inductor backend requires MSVC cl compiler not available in CI)
  • Skip TestLazyMemmapStorageCleanup class on Windows (file locking prevents memmap file deletion/cleanup)
  • Skip editable install test on Windows (file locking prevents .pyd file overwrite during re-install)

Test plan

  • Windows optional deps CI should pass after this change
  • All skipped tests remain active on Linux/macOS

Fixes failing tests from run https://github.com/pytorch/rl/actions/runs/21143115049

Skip tests that fail on Windows due to:
- Missing MSVC compiler for torch.compile inductor backend
- Windows file locking preventing memmap cleanup and editable installs

Tests skipped:
- TestValueFunctionCompile class (compile tests need cl compiler)
- TestLazyMemmapStorageCleanup class (file locking prevents cleanup)
- Editable install test (file locking prevents .pyd overwrite)
@pytorch-bot

pytorch-bot Bot commented Jan 19, 2026

Copy link
Copy Markdown

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3348

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

❌ 1 New Failure, 14 Pending

As of commit f749324 with merge base 0982fcf (image):

NEW FAILURE - The following job has failed:

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 Jan 19, 2026
@vmoens vmoens added bug Something isn't working Tests Incomplete or broken unit tests CI Has to do with CI setup (e.g. wheels & builds, tests...) labels Jan 19, 2026
@github-actions

Copy link
Copy Markdown
Contributor

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

Total Benchmarks: 148. 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] 81.0630μs 79.1344μs 12.6367 KOps/s 12.6936 KOps/s $\color{#d91a1a}-0.45\%$
test_tensor_to_bytestream_speed[torch.save] 0.1411ms 0.1384ms 7.2244 KOps/s 7.3445 KOps/s $\color{#d91a1a}-1.63\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1014s 0.1012s 9.8795 Ops/s 9.7983 Ops/s $\color{#35bf28}+0.83\%$
test_tensor_to_bytestream_speed[numpy] 2.4851μs 2.4772μs 403.6813 KOps/s 417.8722 KOps/s $\color{#d91a1a}-3.40\%$
test_tensor_to_bytestream_speed[safetensors] 40.1099μs 39.8389μs 25.1011 KOps/s 25.7596 KOps/s $\color{#d91a1a}-2.56\%$
test_simple 0.8928s 0.8010s 1.2484 Ops/s 1.2191 Ops/s $\color{#35bf28}+2.40\%$
test_transformed 1.5028s 1.4165s 0.7060 Ops/s 0.7061 Ops/s $\color{#d91a1a}-0.02\%$
test_serial 2.3584s 2.2676s 0.4410 Ops/s 0.4417 Ops/s $\color{#d91a1a}-0.17\%$
test_parallel 2.1311s 1.9674s 0.5083 Ops/s 0.5062 Ops/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-True-True-True-True] 0.3165ms 43.2125μs 23.1415 KOps/s 22.6878 KOps/s $\color{#35bf28}+2.00\%$
test_step_mdp_speed[True-True-True-True-False] 55.6630μs 24.5632μs 40.7113 KOps/s 41.5043 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[True-True-True-False-True] 65.2450μs 24.6305μs 40.6001 KOps/s 40.9581 KOps/s $\color{#d91a1a}-0.87\%$
test_step_mdp_speed[True-True-True-False-False] 51.8830μs 13.4552μs 74.3205 KOps/s 73.5833 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[True-True-False-True-True] 89.2560μs 47.0470μs 21.2553 KOps/s 21.3832 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[True-True-False-True-False] 63.4440μs 27.1702μs 36.8050 KOps/s 36.4505 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-True-False-False-True] 64.3640μs 27.2351μs 36.7174 KOps/s 36.6716 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[True-True-False-False-False] 52.4630μs 16.2654μs 61.4800 KOps/s 60.8673 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[True-False-True-True-True] 0.1034ms 49.3695μs 20.2554 KOps/s 20.3404 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[True-False-True-True-False] 63.1140μs 29.7175μs 33.6502 KOps/s 34.1011 KOps/s $\color{#d91a1a}-1.32\%$
test_step_mdp_speed[True-False-True-False-True] 66.6950μs 27.6421μs 36.1767 KOps/s 37.4639 KOps/s $\color{#d91a1a}-3.44\%$
test_step_mdp_speed[True-False-True-False-False] 50.8330μs 16.3875μs 61.0221 KOps/s 61.7490 KOps/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-False-False-True-True] 91.6460μs 52.6178μs 19.0050 KOps/s 19.4729 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[True-False-False-True-False] 67.7740μs 32.1031μs 31.1496 KOps/s 31.2025 KOps/s $\color{#d91a1a}-0.17\%$
test_step_mdp_speed[True-False-False-False-True] 68.7240μs 29.5643μs 33.8246 KOps/s 34.2168 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[True-False-False-False-False] 47.3630μs 18.9863μs 52.6697 KOps/s 53.1772 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[False-True-True-True-True] 96.5260μs 49.6979μs 20.1216 KOps/s 20.2693 KOps/s $\color{#d91a1a}-0.73\%$
test_step_mdp_speed[False-True-True-True-False] 72.4250μs 29.8790μs 33.4683 KOps/s 34.3439 KOps/s $\color{#d91a1a}-2.55\%$
test_step_mdp_speed[False-True-True-False-True] 65.0740μs 31.2962μs 31.9527 KOps/s 32.5268 KOps/s $\color{#d91a1a}-1.77\%$
test_step_mdp_speed[False-True-True-False-False] 54.6930μs 17.9405μs 55.7398 KOps/s 56.3650 KOps/s $\color{#d91a1a}-1.11\%$
test_step_mdp_speed[False-True-False-True-True] 2.6959ms 52.7245μs 18.9665 KOps/s 19.7699 KOps/s $\color{#d91a1a}-4.06\%$
test_step_mdp_speed[False-True-False-True-False] 72.9640μs 32.8708μs 30.4222 KOps/s 31.4013 KOps/s $\color{#d91a1a}-3.12\%$
test_step_mdp_speed[False-True-False-False-True] 69.7040μs 34.0796μs 29.3431 KOps/s 30.2836 KOps/s $\color{#d91a1a}-3.11\%$
test_step_mdp_speed[False-True-False-False-False] 55.4130μs 20.6896μs 48.3335 KOps/s 49.0853 KOps/s $\color{#d91a1a}-1.53\%$
test_step_mdp_speed[False-False-True-True-True] 88.8250μs 55.1627μs 18.1282 KOps/s 18.4779 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[False-False-True-True-False] 71.3940μs 35.3972μs 28.2508 KOps/s 28.5239 KOps/s $\color{#d91a1a}-0.96\%$
test_step_mdp_speed[False-False-True-False-True] 72.0150μs 33.4192μs 29.9229 KOps/s 29.5351 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[False-False-True-False-False] 57.6440μs 20.5618μs 48.6338 KOps/s 48.5470 KOps/s $\color{#35bf28}+0.18\%$
test_step_mdp_speed[False-False-False-True-True] 0.1210ms 57.1149μs 17.5086 KOps/s 17.7146 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[False-False-False-True-False] 77.9150μs 38.2778μs 26.1248 KOps/s 26.6778 KOps/s $\color{#d91a1a}-2.07\%$
test_step_mdp_speed[False-False-False-False-True] 69.6540μs 35.6743μs 28.0314 KOps/s 28.2631 KOps/s $\color{#d91a1a}-0.82\%$
test_step_mdp_speed[False-False-False-False-False] 54.7240μs 23.1023μs 43.2857 KOps/s 43.5746 KOps/s $\color{#d91a1a}-0.66\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8461s 0.7758s 1.2890 Ops/s 1.3266 Ops/s $\color{#d91a1a}-2.83\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7161s 0.6203s 1.6122 Ops/s 1.6154 Ops/s $\color{#d91a1a}-0.20\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7204s 1.6413s 0.6093 Ops/s 0.6127 Ops/s $\color{#d91a1a}-0.56\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.4947s 1.4175s 0.7055 Ops/s 0.7031 Ops/s $\color{#35bf28}+0.34\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9801s 1.8966s 0.5273 Ops/s 0.5311 Ops/s $\color{#d91a1a}-0.73\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7306s 1.6564s 0.6037 Ops/s 0.6030 Ops/s $\color{#35bf28}+0.12\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.6791s 4.5929s 0.2177 Ops/s 0.2178 Ops/s $\color{#d91a1a}-0.05\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4763s 4.4245s 0.2260 Ops/s 0.2260 Ops/s $+0.01\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0149s 1.9287s 0.5185 Ops/s 0.5151 Ops/s $\color{#35bf28}+0.66\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7558s 1.6663s 0.6001 Ops/s 0.6063 Ops/s $\color{#d91a1a}-1.02\%$
test_values[generalized_advantage_estimate-True-True] 20.2010ms 19.4933ms 51.2997 Ops/s 50.7938 Ops/s $\color{#35bf28}+1.00\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1463s 3.8343ms 260.8052 Ops/s 266.7852 Ops/s $\color{#d91a1a}-2.24\%$
test_values[td0_return_estimate-False-False] 0.1078ms 81.7353μs 12.2346 KOps/s 12.3900 KOps/s $\color{#d91a1a}-1.25\%$
test_values[td1_return_estimate-False-False] 48.5759ms 46.6465ms 21.4378 Ops/s 21.4910 Ops/s $\color{#d91a1a}-0.25\%$
test_values[vec_td1_return_estimate-False-False] 1.3472ms 1.0702ms 934.3767 Ops/s 931.0918 Ops/s $\color{#35bf28}+0.35\%$
test_values[td_lambda_return_estimate-True-False] 80.5267ms 77.6751ms 12.8741 Ops/s 13.0196 Ops/s $\color{#d91a1a}-1.12\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2904ms 1.0660ms 938.1190 Ops/s 933.5376 Ops/s $\color{#35bf28}+0.49\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 21.2387ms 20.9580ms 47.7144 Ops/s 50.6499 Ops/s $\textbf{\color{#d91a1a}-5.80\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0133ms 0.7361ms 1.3584 KOps/s 1.3543 KOps/s $\color{#35bf28}+0.31\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7063ms 0.6640ms 1.5061 KOps/s 1.4987 KOps/s $\color{#35bf28}+0.49\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5282ms 1.4733ms 678.7366 Ops/s 676.7971 Ops/s $\color{#35bf28}+0.29\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7403ms 0.6802ms 1.4702 KOps/s 1.4614 KOps/s $\color{#35bf28}+0.60\%$
test_dqn_speed[False-None] 1.5890ms 1.5118ms 661.4621 Ops/s 662.6227 Ops/s $\color{#d91a1a}-0.18\%$
test_dqn_speed[False-backward] 2.2099ms 2.1387ms 467.5694 Ops/s 465.6589 Ops/s $\color{#35bf28}+0.41\%$
test_dqn_speed[True-None] 1.5195ms 0.5779ms 1.7304 KOps/s 1.7137 KOps/s $\color{#35bf28}+0.98\%$
test_dqn_speed[True-backward] 1.1445ms 1.0965ms 911.9899 Ops/s 825.4463 Ops/s $\textbf{\color{#35bf28}+10.48\%}$
test_dqn_speed[reduce-overhead-None] 0.6494ms 0.5833ms 1.7145 KOps/s 1.6498 KOps/s $\color{#35bf28}+3.92\%$
test_ddpg_speed[False-None] 3.2249ms 2.8622ms 349.3787 Ops/s 348.6444 Ops/s $\color{#35bf28}+0.21\%$
test_ddpg_speed[False-backward] 4.5152ms 4.1093ms 243.3494 Ops/s 235.9054 Ops/s $\color{#35bf28}+3.16\%$
test_ddpg_speed[True-None] 1.4790ms 1.3414ms 745.4685 Ops/s 747.0083 Ops/s $\color{#d91a1a}-0.21\%$
test_ddpg_speed[True-backward] 2.4483ms 2.4022ms 416.2887 Ops/s 389.8676 Ops/s $\textbf{\color{#35bf28}+6.78\%}$
test_ddpg_speed[reduce-overhead-None] 1.4397ms 1.3574ms 736.6856 Ops/s 730.8219 Ops/s $\color{#35bf28}+0.80\%$
test_sac_speed[False-None] 8.7290ms 8.1830ms 122.2041 Ops/s 121.6293 Ops/s $\color{#35bf28}+0.47\%$
test_sac_speed[False-backward] 11.5679ms 11.1150ms 89.9686 Ops/s 87.2977 Ops/s $\color{#35bf28}+3.06\%$
test_sac_speed[True-None] 1.9270ms 1.8443ms 542.2236 Ops/s 540.9828 Ops/s $\color{#35bf28}+0.23\%$
test_sac_speed[True-backward] 3.5173ms 3.4569ms 289.2776 Ops/s 282.2763 Ops/s $\color{#35bf28}+2.48\%$
test_sac_speed[reduce-overhead-None] 18.1999ms 10.4364ms 95.8189 Ops/s 95.0323 Ops/s $\color{#35bf28}+0.83\%$
test_redq_deprec_speed[False-None] 9.7317ms 9.1384ms 109.4287 Ops/s 108.3907 Ops/s $\color{#35bf28}+0.96\%$
test_redq_deprec_speed[False-backward] 12.7007ms 12.1311ms 82.4328 Ops/s 81.7460 Ops/s $\color{#35bf28}+0.84\%$
test_redq_deprec_speed[True-None] 2.6237ms 2.5457ms 392.8171 Ops/s 378.0741 Ops/s $\color{#35bf28}+3.90\%$
test_redq_deprec_speed[True-backward] 4.2946ms 4.1604ms 240.3610 Ops/s 224.6862 Ops/s $\textbf{\color{#35bf28}+6.98\%}$
test_redq_deprec_speed[reduce-overhead-None] 15.2267ms 9.5028ms 105.2317 Ops/s 123.8325 Ops/s $\textbf{\color{#d91a1a}-15.02\%}$
test_td3_speed[False-None] 8.1561ms 8.0642ms 124.0051 Ops/s 121.8607 Ops/s $\color{#35bf28}+1.76\%$
test_td3_speed[False-backward] 10.9600ms 10.4262ms 95.9118 Ops/s 93.2764 Ops/s $\color{#35bf28}+2.83\%$
test_td3_speed[True-None] 1.7957ms 1.7117ms 584.2119 Ops/s 582.6377 Ops/s $\color{#35bf28}+0.27\%$
test_td3_speed[True-backward] 3.2794ms 3.1792ms 314.5435 Ops/s 298.1701 Ops/s $\textbf{\color{#35bf28}+5.49\%}$
test_td3_speed[reduce-overhead-None] 54.3034ms 23.2034ms 43.0971 Ops/s 42.1742 Ops/s $\color{#35bf28}+2.19\%$
test_cql_speed[False-None] 17.2063ms 16.9071ms 59.1467 Ops/s 58.8121 Ops/s $\color{#35bf28}+0.57\%$
test_cql_speed[False-backward] 22.5802ms 22.0323ms 45.3879 Ops/s 44.4316 Ops/s $\color{#35bf28}+2.15\%$
test_cql_speed[True-None] 3.6045ms 3.4077ms 293.4554 Ops/s 288.3131 Ops/s $\color{#35bf28}+1.78\%$
test_cql_speed[True-backward] 5.8917ms 5.5673ms 179.6209 Ops/s 173.8312 Ops/s $\color{#35bf28}+3.33\%$
test_cql_speed[reduce-overhead-None] 17.8506ms 11.4023ms 87.7013 Ops/s 86.4121 Ops/s $\color{#35bf28}+1.49\%$
test_a2c_speed[False-None] 3.8298ms 3.1641ms 316.0469 Ops/s 314.2734 Ops/s $\color{#35bf28}+0.56\%$
test_a2c_speed[False-backward] 6.4649ms 6.0454ms 165.4144 Ops/s 158.1456 Ops/s $\color{#35bf28}+4.60\%$
test_a2c_speed[True-None] 1.4417ms 1.3390ms 746.8460 Ops/s 742.1067 Ops/s $\color{#35bf28}+0.64\%$
test_a2c_speed[True-backward] 3.0543ms 2.9951ms 333.8787 Ops/s 314.7488 Ops/s $\textbf{\color{#35bf28}+6.08\%}$
test_a2c_speed[reduce-overhead-None] 1.0264ms 0.9441ms 1.0592 KOps/s 1.0565 KOps/s $\color{#35bf28}+0.26\%$
test_ppo_speed[False-None] 3.8137ms 3.7416ms 267.2627 Ops/s 265.2902 Ops/s $\color{#35bf28}+0.74\%$
test_ppo_speed[False-backward] 7.2021ms 6.7801ms 147.4915 Ops/s 141.2905 Ops/s $\color{#35bf28}+4.39\%$
test_ppo_speed[True-None] 1.4690ms 1.3953ms 716.7142 Ops/s 706.3785 Ops/s $\color{#35bf28}+1.46\%$
test_ppo_speed[True-backward] 3.4051ms 3.3092ms 302.1835 Ops/s 309.7837 Ops/s $\color{#d91a1a}-2.45\%$
test_ppo_speed[reduce-overhead-None] 1.0870ms 1.0019ms 998.1214 Ops/s 971.6488 Ops/s $\color{#35bf28}+2.72\%$
test_reinforce_speed[False-None] 2.3089ms 2.2312ms 448.1894 Ops/s 442.8777 Ops/s $\color{#35bf28}+1.20\%$
test_reinforce_speed[False-backward] 3.5817ms 3.2344ms 309.1732 Ops/s 294.8425 Ops/s $\color{#35bf28}+4.86\%$
test_reinforce_speed[True-None] 1.8301ms 1.2592ms 794.1433 Ops/s 768.7674 Ops/s $\color{#35bf28}+3.30\%$
test_reinforce_speed[True-backward] 3.6633ms 3.1445ms 318.0189 Ops/s 330.9163 Ops/s $\color{#d91a1a}-3.90\%$
test_reinforce_speed[reduce-overhead-None] 0.4620s 9.8500ms 101.5230 Ops/s 97.1586 Ops/s $\color{#35bf28}+4.49\%$
test_iql_speed[False-None] 9.9313ms 9.2584ms 108.0106 Ops/s 106.7064 Ops/s $\color{#35bf28}+1.22\%$
test_iql_speed[False-backward] 13.5946ms 13.2166ms 75.6625 Ops/s 76.3325 Ops/s $\color{#d91a1a}-0.88\%$
test_iql_speed[True-None] 2.4704ms 2.2148ms 451.5144 Ops/s 446.9981 Ops/s $\color{#35bf28}+1.01\%$
test_iql_speed[True-backward] 5.3861ms 5.0262ms 198.9591 Ops/s 195.5853 Ops/s $\color{#35bf28}+1.73\%$
test_iql_speed[reduce-overhead-None] 0.7088s 12.1690ms 82.1761 Ops/s 100.6492 Ops/s $\textbf{\color{#d91a1a}-18.35\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.3415ms 5.8840ms 169.9518 Ops/s 169.7739 Ops/s $\color{#35bf28}+0.10\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6988ms 0.3186ms 3.1383 KOps/s 3.1881 KOps/s $\color{#d91a1a}-1.56\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5791ms 0.3598ms 2.7796 KOps/s 3.3724 KOps/s $\textbf{\color{#d91a1a}-17.58\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9586ms 5.6414ms 177.2597 Ops/s 177.0215 Ops/s $\color{#35bf28}+0.13\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5377ms 0.2723ms 3.6722 KOps/s 3.2118 KOps/s $\textbf{\color{#35bf28}+14.33\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6821ms 0.2555ms 3.9135 KOps/s 3.2099 KOps/s $\textbf{\color{#35bf28}+21.92\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6260ms 1.3649ms 732.6406 Ops/s 801.9714 Ops/s $\textbf{\color{#d91a1a}-8.65\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.3297ms 1.1358ms 880.4092 Ops/s 852.4963 Ops/s $\color{#35bf28}+3.27\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.7895ms 5.9607ms 167.7653 Ops/s 172.2545 Ops/s $\color{#d91a1a}-2.61\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.2069ms 0.4394ms 2.2756 KOps/s 2.3861 KOps/s $\color{#d91a1a}-4.63\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8117ms 0.4340ms 2.3042 KOps/s 2.2427 KOps/s $\color{#35bf28}+2.74\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8771ms 5.7485ms 173.9573 Ops/s 176.1926 Ops/s $\color{#d91a1a}-1.27\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0136ms 0.3452ms 2.8973 KOps/s 2.7913 KOps/s $\color{#35bf28}+3.80\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5131ms 0.2596ms 3.8519 KOps/s 2.9084 KOps/s $\textbf{\color{#35bf28}+32.44\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8121ms 5.6055ms 178.3957 Ops/s 177.9474 Ops/s $\color{#35bf28}+0.25\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6637ms 0.3089ms 3.2373 KOps/s 3.2983 KOps/s $\color{#d91a1a}-1.85\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4443ms 0.2577ms 3.8803 KOps/s 3.3116 KOps/s $\textbf{\color{#35bf28}+17.17\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.9342ms 5.8234ms 171.7205 Ops/s 173.6893 Ops/s $\color{#d91a1a}-1.13\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9854ms 0.4754ms 2.1034 KOps/s 2.1648 KOps/s $\color{#d91a1a}-2.84\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8481ms 0.5171ms 1.9338 KOps/s 2.2389 KOps/s $\textbf{\color{#d91a1a}-13.63\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4331ms 4.9310ms 202.7986 Ops/s 49.6953 Ops/s $\textbf{\color{#35bf28}+308.08\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.0047ms 2.2176ms 450.9365 Ops/s 544.6509 Ops/s $\textbf{\color{#d91a1a}-17.21\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.3987ms 1.2581ms 794.8184 Ops/s 1.0966 KOps/s $\textbf{\color{#d91a1a}-27.52\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5831s 16.5849ms 60.2957 Ops/s 197.0529 Ops/s $\textbf{\color{#d91a1a}-69.40\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 3.8811ms 1.7293ms 578.2702 Ops/s 518.2282 Ops/s $\textbf{\color{#35bf28}+11.59\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.2181ms 1.1883ms 841.5383 Ops/s 732.3196 Ops/s $\textbf{\color{#35bf28}+14.91\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.7050ms 5.1766ms 193.1784 Ops/s 191.1399 Ops/s $\color{#35bf28}+1.07\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 13.0212ms 2.0682ms 483.5088 Ops/s 486.4222 Ops/s $\color{#d91a1a}-0.60\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.4237ms 1.0829ms 923.4256 Ops/s 898.5311 Ops/s $\color{#35bf28}+2.77\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.8000ms 33.6697ms 29.7003 Ops/s 29.3593 Ops/s $\color{#35bf28}+1.16\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.7136ms 17.1961ms 58.1526 Ops/s 57.6521 Ops/s $\color{#35bf28}+0.87\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 38.2216ms 34.7518ms 28.7755 Ops/s 28.5400 Ops/s $\color{#35bf28}+0.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.6423ms 17.5263ms 57.0571 Ops/s 55.3298 Ops/s $\color{#35bf28}+3.12\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 38.8056ms 36.6498ms 27.2852 Ops/s 27.2364 Ops/s $\color{#35bf28}+0.18\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.1206ms 18.7439ms 53.3507 Ops/s 52.4172 Ops/s $\color{#35bf28}+1.78\%$

@github-actions

Copy link
Copy Markdown
Contributor

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

Total Benchmarks: 153. Improved: $\large\color{#35bf28}16$. 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] 84.9644μs 82.5868μs 12.1085 KOps/s 12.3217 KOps/s $\color{#d91a1a}-1.73\%$
test_tensor_to_bytestream_speed[torch.save] 0.1444ms 0.1432ms 6.9850 KOps/s 7.0785 KOps/s $\color{#d91a1a}-1.32\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1276s 0.1269s 7.8800 Ops/s 8.0282 Ops/s $\color{#d91a1a}-1.85\%$
test_tensor_to_bytestream_speed[numpy] 2.5108μs 2.4989μs 400.1777 KOps/s 405.9687 KOps/s $\color{#d91a1a}-1.43\%$
test_tensor_to_bytestream_speed[safetensors] 40.3756μs 40.1936μs 24.8796 KOps/s 25.2509 KOps/s $\color{#d91a1a}-1.47\%$
test_simple 0.5531s 0.5514s 1.8136 Ops/s 1.7345 Ops/s $\color{#35bf28}+4.56\%$
test_transformed 1.2471s 1.1553s 0.8656 Ops/s 0.8594 Ops/s $\color{#35bf28}+0.72\%$
test_serial 1.6785s 1.6731s 0.5977 Ops/s 0.5845 Ops/s $\color{#35bf28}+2.25\%$
test_parallel 1.3573s 1.2147s 0.8233 Ops/s 0.8604 Ops/s $\color{#d91a1a}-4.32\%$
test_step_mdp_speed[True-True-True-True-True] 0.3141ms 45.7066μs 21.8787 KOps/s 22.3858 KOps/s $\color{#d91a1a}-2.27\%$
test_step_mdp_speed[True-True-True-True-False] 50.3520μs 25.5111μs 39.1986 KOps/s 40.5855 KOps/s $\color{#d91a1a}-3.42\%$
test_step_mdp_speed[True-True-True-False-True] 58.9710μs 25.6296μs 39.0174 KOps/s 40.5813 KOps/s $\color{#d91a1a}-3.85\%$
test_step_mdp_speed[True-True-True-False-False] 40.8710μs 13.9360μs 71.7566 KOps/s 73.3374 KOps/s $\color{#d91a1a}-2.16\%$
test_step_mdp_speed[True-True-False-True-True] 80.6220μs 48.2555μs 20.7230 KOps/s 20.8107 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[True-True-False-True-False] 66.2120μs 27.8552μs 35.8999 KOps/s 35.9935 KOps/s $\color{#d91a1a}-0.26\%$
test_step_mdp_speed[True-True-False-False-True] 60.1910μs 28.0505μs 35.6500 KOps/s 36.2598 KOps/s $\color{#d91a1a}-1.68\%$
test_step_mdp_speed[True-True-False-False-False] 46.1610μs 16.8116μs 59.4829 KOps/s 61.4787 KOps/s $\color{#d91a1a}-3.25\%$
test_step_mdp_speed[True-False-True-True-True] 83.2520μs 51.6869μs 19.3473 KOps/s 19.8882 KOps/s $\color{#d91a1a}-2.72\%$
test_step_mdp_speed[True-False-True-True-False] 70.6010μs 30.6994μs 32.5739 KOps/s 32.6932 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[True-False-True-False-True] 57.3210μs 28.4114μs 35.1971 KOps/s 35.6840 KOps/s $\color{#d91a1a}-1.36\%$
test_step_mdp_speed[True-False-True-False-False] 76.1620μs 16.6443μs 60.0806 KOps/s 60.4437 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[True-False-False-True-True] 0.1285ms 53.3716μs 18.7366 KOps/s 18.7916 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[True-False-False-True-False] 68.7010μs 33.3009μs 30.0293 KOps/s 29.9444 KOps/s $\color{#35bf28}+0.28\%$
test_step_mdp_speed[True-False-False-False-True] 73.1910μs 30.7300μs 32.5415 KOps/s 32.8845 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[True-False-False-False-False] 45.1410μs 19.4007μs 51.5446 KOps/s 51.8893 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[False-True-True-True-True] 85.0610μs 50.7773μs 19.6938 KOps/s 19.6698 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[False-True-True-True-False] 65.6620μs 30.6095μs 32.6696 KOps/s 32.7600 KOps/s $\color{#d91a1a}-0.28\%$
test_step_mdp_speed[False-True-True-False-True] 63.4010μs 31.7181μs 31.5278 KOps/s 31.7872 KOps/s $\color{#d91a1a}-0.82\%$
test_step_mdp_speed[False-True-True-False-False] 58.1910μs 18.1475μs 55.1041 KOps/s 54.7420 KOps/s $\color{#35bf28}+0.66\%$
test_step_mdp_speed[False-True-False-True-True] 2.7384ms 53.5382μs 18.6782 KOps/s 18.8160 KOps/s $\color{#d91a1a}-0.73\%$
test_step_mdp_speed[False-True-False-True-False] 64.4210μs 33.4039μs 29.9366 KOps/s 29.6764 KOps/s $\color{#35bf28}+0.88\%$
test_step_mdp_speed[False-True-False-False-True] 71.5610μs 34.3017μs 29.1531 KOps/s 29.1460 KOps/s $\color{#35bf28}+0.02\%$
test_step_mdp_speed[False-True-False-False-False] 60.4020μs 20.8278μs 48.0126 KOps/s 47.7071 KOps/s $\color{#35bf28}+0.64\%$
test_step_mdp_speed[False-False-True-True-True] 94.0720μs 56.2554μs 17.7761 KOps/s 17.6296 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-False-True-True-False] 72.6310μs 36.1319μs 27.6764 KOps/s 27.1646 KOps/s $\color{#35bf28}+1.88\%$
test_step_mdp_speed[False-False-True-False-True] 69.2110μs 33.7994μs 29.5863 KOps/s 29.0438 KOps/s $\color{#35bf28}+1.87\%$
test_step_mdp_speed[False-False-True-False-False] 53.6110μs 20.7138μs 48.2771 KOps/s 47.6941 KOps/s $\color{#35bf28}+1.22\%$
test_step_mdp_speed[False-False-False-True-True] 86.3720μs 57.9424μs 17.2585 KOps/s 17.0097 KOps/s $\color{#35bf28}+1.46\%$
test_step_mdp_speed[False-False-False-True-False] 66.1310μs 38.9925μs 25.6460 KOps/s 25.8393 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-False-False-False-True] 66.7020μs 36.3329μs 27.5233 KOps/s 27.7745 KOps/s $\color{#d91a1a}-0.90\%$
test_step_mdp_speed[False-False-False-False-False] 48.3510μs 23.5746μs 42.4185 KOps/s 42.8690 KOps/s $\color{#d91a1a}-1.05\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8720s 0.7718s 1.2957 Ops/s 1.2869 Ops/s $\color{#35bf28}+0.69\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7348s 0.6362s 1.5719 Ops/s 1.5580 Ops/s $\color{#35bf28}+0.90\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7591s 1.6775s 0.5961 Ops/s 0.5897 Ops/s $\color{#35bf28}+1.09\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5323s 1.4533s 0.6881 Ops/s 0.6798 Ops/s $\color{#35bf28}+1.22\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0101s 1.9341s 0.5170 Ops/s 0.5155 Ops/s $\color{#35bf28}+0.30\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7858s 1.7095s 0.5850 Ops/s 0.5810 Ops/s $\color{#35bf28}+0.67\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.7701s 4.6682s 0.2142 Ops/s 0.2118 Ops/s $\color{#35bf28}+1.15\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4868s 4.4437s 0.2250 Ops/s 0.2218 Ops/s $\color{#35bf28}+1.46\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1152s 1.9962s 0.5009 Ops/s 0.5120 Ops/s $\color{#d91a1a}-2.16\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.8030s 1.6849s 0.5935 Ops/s 0.6040 Ops/s $\color{#d91a1a}-1.75\%$
test_values[generalized_advantage_estimate-True-True] 11.0693ms 10.8659ms 92.0311 Ops/s 91.3378 Ops/s $\color{#35bf28}+0.76\%$
test_values[vec_generalized_advantage_estimate-True-True] 20.1899ms 17.9341ms 55.7596 Ops/s 55.7138 Ops/s $\color{#35bf28}+0.08\%$
test_values[td0_return_estimate-False-False] 0.2632ms 0.1299ms 7.6974 KOps/s 7.7741 KOps/s $\color{#d91a1a}-0.99\%$
test_values[td1_return_estimate-False-False] 29.7578ms 29.2846ms 34.1476 Ops/s 33.7034 Ops/s $\color{#35bf28}+1.32\%$
test_values[vec_td1_return_estimate-False-False] 18.2765ms 17.9471ms 55.7193 Ops/s 55.4640 Ops/s $\color{#35bf28}+0.46\%$
test_values[td_lambda_return_estimate-True-False] 44.0305ms 43.3248ms 23.0815 Ops/s 22.5403 Ops/s $\color{#35bf28}+2.40\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.3521ms 17.9260ms 55.7848 Ops/s 55.4788 Ops/s $\color{#35bf28}+0.55\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.7891ms 9.6659ms 103.4561 Ops/s 102.4398 Ops/s $\color{#35bf28}+0.99\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7435ms 1.5471ms 646.3884 Ops/s 651.6535 Ops/s $\color{#d91a1a}-0.81\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6428ms 0.4373ms 2.2866 KOps/s 2.3024 KOps/s $\color{#d91a1a}-0.69\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.1180ms 34.1540ms 29.2792 Ops/s 29.0059 Ops/s $\color{#35bf28}+0.94\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.1380ms 1.7605ms 568.0311 Ops/s 564.7481 Ops/s $\color{#35bf28}+0.58\%$
test_dqn_speed[False-None] 1.7010ms 1.4260ms 701.2563 Ops/s 697.6540 Ops/s $\color{#35bf28}+0.52\%$
test_dqn_speed[False-backward] 2.0337ms 1.9568ms 511.0489 Ops/s 510.9788 Ops/s $\color{#35bf28}+0.01\%$
test_dqn_speed[True-None] 0.9493ms 0.5372ms 1.8615 KOps/s 1.8468 KOps/s $\color{#35bf28}+0.80\%$
test_dqn_speed[True-backward] 1.1049ms 1.0109ms 989.1870 Ops/s 826.4449 Ops/s $\textbf{\color{#35bf28}+19.69\%}$
test_dqn_speed[reduce-overhead-None] 0.6528ms 0.5304ms 1.8854 KOps/s 1.8350 KOps/s $\color{#35bf28}+2.75\%$
test_ddpg_speed[False-None] 3.2734ms 2.8724ms 348.1401 Ops/s 343.5552 Ops/s $\color{#35bf28}+1.33\%$
test_ddpg_speed[False-backward] 4.5592ms 4.1646ms 240.1206 Ops/s 241.2133 Ops/s $\color{#d91a1a}-0.45\%$
test_ddpg_speed[True-None] 1.7613ms 1.3997ms 714.4521 Ops/s 704.2398 Ops/s $\color{#35bf28}+1.45\%$
test_ddpg_speed[True-backward] 2.4476ms 2.4005ms 416.5860 Ops/s 365.9346 Ops/s $\textbf{\color{#35bf28}+13.84\%}$
test_ddpg_speed[reduce-overhead-None] 1.6046ms 1.3967ms 715.9493 Ops/s 702.3140 Ops/s $\color{#35bf28}+1.94\%$
test_sac_speed[False-None] 8.9117ms 8.1780ms 122.2797 Ops/s 121.1419 Ops/s $\color{#35bf28}+0.94\%$
test_sac_speed[False-backward] 11.9572ms 11.4701ms 87.1834 Ops/s 86.7226 Ops/s $\color{#35bf28}+0.53\%$
test_sac_speed[True-None] 3.1667ms 2.2191ms 450.6374 Ops/s 467.9554 Ops/s $\color{#d91a1a}-3.70\%$
test_sac_speed[True-backward] 4.2657ms 4.0651ms 245.9992 Ops/s 207.8290 Ops/s $\textbf{\color{#35bf28}+18.37\%}$
test_sac_speed[reduce-overhead-None] 2.5371ms 2.1609ms 462.7725 Ops/s 449.4726 Ops/s $\color{#35bf28}+2.96\%$
test_redq_speed[False-None] 10.9482ms 10.4038ms 96.1183 Ops/s 95.1505 Ops/s $\color{#35bf28}+1.02\%$
test_redq_speed[False-backward] 19.1394ms 17.9828ms 55.6087 Ops/s 56.8820 Ops/s $\color{#d91a1a}-2.24\%$
test_redq_speed[True-None] 4.6186ms 4.4278ms 225.8465 Ops/s 233.2065 Ops/s $\color{#d91a1a}-3.16\%$
test_redq_speed[True-backward] 10.1612ms 9.7801ms 102.2485 Ops/s 94.1583 Ops/s $\textbf{\color{#35bf28}+8.59\%}$
test_redq_speed[reduce-overhead-None] 4.7249ms 4.3624ms 229.2330 Ops/s 233.1887 Ops/s $\color{#d91a1a}-1.70\%$
test_redq_deprec_speed[False-None] 11.6105ms 11.1712ms 89.5162 Ops/s 90.4363 Ops/s $\color{#d91a1a}-1.02\%$
test_redq_deprec_speed[False-backward] 16.4831ms 16.1045ms 62.0945 Ops/s 63.1579 Ops/s $\color{#d91a1a}-1.68\%$
test_redq_deprec_speed[True-None] 4.9190ms 3.6967ms 270.5131 Ops/s 262.3462 Ops/s $\color{#35bf28}+3.11\%$
test_redq_deprec_speed[True-backward] 7.8646ms 7.6323ms 131.0226 Ops/s 126.1118 Ops/s $\color{#35bf28}+3.89\%$
test_redq_deprec_speed[reduce-overhead-None] 4.5576ms 3.6644ms 272.8932 Ops/s 272.6403 Ops/s $\color{#35bf28}+0.09\%$
test_td3_speed[False-None] 8.3913ms 8.1463ms 122.7553 Ops/s 121.7901 Ops/s $\color{#35bf28}+0.79\%$
test_td3_speed[False-backward] 11.4639ms 11.0929ms 90.1474 Ops/s 90.0368 Ops/s $\color{#35bf28}+0.12\%$
test_td3_speed[True-None] 1.8888ms 1.8446ms 542.1169 Ops/s 541.4532 Ops/s $\color{#35bf28}+0.12\%$
test_td3_speed[True-backward] 3.7706ms 3.6574ms 273.4208 Ops/s 238.4629 Ops/s $\textbf{\color{#35bf28}+14.66\%}$
test_td3_speed[reduce-overhead-None] 1.8639ms 1.8045ms 554.1687 Ops/s 550.7789 Ops/s $\color{#35bf28}+0.62\%$
test_cql_speed[False-None] 29.5828ms 26.5785ms 37.6243 Ops/s 38.1848 Ops/s $\color{#d91a1a}-1.47\%$
test_cql_speed[False-backward] 38.8570ms 35.8422ms 27.9000 Ops/s 27.7029 Ops/s $\color{#35bf28}+0.71\%$
test_cql_speed[True-None] 13.2384ms 12.5121ms 79.9229 Ops/s 79.9674 Ops/s $\color{#d91a1a}-0.06\%$
test_cql_speed[True-backward] 19.2022ms 18.6879ms 53.5105 Ops/s 53.4769 Ops/s $\color{#35bf28}+0.06\%$
test_cql_speed[reduce-overhead-None] 15.5018ms 12.5397ms 79.7468 Ops/s 81.2608 Ops/s $\color{#d91a1a}-1.86\%$
test_a2c_speed[False-None] 6.0370ms 5.5149ms 181.3278 Ops/s 183.8332 Ops/s $\color{#d91a1a}-1.36\%$
test_a2c_speed[False-backward] 12.3796ms 11.9752ms 83.5061 Ops/s 83.8211 Ops/s $\color{#d91a1a}-0.38\%$
test_a2c_speed[True-None] 4.0822ms 3.7239ms 268.5391 Ops/s 269.0357 Ops/s $\color{#d91a1a}-0.18\%$
test_a2c_speed[True-backward] 8.7691ms 8.5659ms 116.7418 Ops/s 118.2043 Ops/s $\color{#d91a1a}-1.24\%$
test_a2c_speed[reduce-overhead-None] 3.9733ms 3.6954ms 270.6062 Ops/s 267.7255 Ops/s $\color{#35bf28}+1.08\%$
test_ppo_speed[False-None] 6.4246ms 5.9929ms 166.8645 Ops/s 168.3579 Ops/s $\color{#d91a1a}-0.89\%$
test_ppo_speed[False-backward] 12.9552ms 12.6659ms 78.9521 Ops/s 79.5096 Ops/s $\color{#d91a1a}-0.70\%$
test_ppo_speed[True-None] 3.8574ms 3.6451ms 274.3408 Ops/s 270.4796 Ops/s $\color{#35bf28}+1.43\%$
test_ppo_speed[True-backward] 8.8148ms 8.4548ms 118.2760 Ops/s 113.2785 Ops/s $\color{#35bf28}+4.41\%$
test_ppo_speed[reduce-overhead-None] 4.3159ms 3.6642ms 272.9123 Ops/s 278.5034 Ops/s $\color{#d91a1a}-2.01\%$
test_reinforce_speed[False-None] 5.0522ms 4.6612ms 214.5374 Ops/s 223.2031 Ops/s $\color{#d91a1a}-3.88\%$
test_reinforce_speed[False-backward] 8.0282ms 7.4204ms 134.7636 Ops/s 136.1559 Ops/s $\color{#d91a1a}-1.02\%$
test_reinforce_speed[True-None] 3.2682ms 2.8234ms 354.1821 Ops/s 351.9435 Ops/s $\color{#35bf28}+0.64\%$
test_reinforce_speed[True-backward] 8.0757ms 7.7824ms 128.4945 Ops/s 119.5797 Ops/s $\textbf{\color{#35bf28}+7.46\%}$
test_reinforce_speed[reduce-overhead-None] 3.2011ms 2.8495ms 350.9411 Ops/s 317.9917 Ops/s $\textbf{\color{#35bf28}+10.36\%}$
test_iql_speed[False-None] 23.1625ms 20.3625ms 49.1100 Ops/s 50.0528 Ops/s $\color{#d91a1a}-1.88\%$
test_iql_speed[False-backward] 31.1906ms 30.6043ms 32.6751 Ops/s 32.6673 Ops/s $\color{#35bf28}+0.02\%$
test_iql_speed[True-None] 8.8261ms 8.4938ms 117.7324 Ops/s 114.7345 Ops/s $\color{#35bf28}+2.61\%$
test_iql_speed[True-backward] 17.2937ms 16.6326ms 60.1229 Ops/s 59.2329 Ops/s $\color{#35bf28}+1.50\%$
test_iql_speed[reduce-overhead-None] 8.8733ms 8.5273ms 117.2706 Ops/s 114.4261 Ops/s $\color{#35bf28}+2.49\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.8267ms 6.1310ms 163.1059 Ops/s 161.8225 Ops/s $\color{#35bf28}+0.79\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7206ms 0.3006ms 3.3268 KOps/s 2.7557 KOps/s $\textbf{\color{#35bf28}+20.72\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7052ms 0.2969ms 3.3685 KOps/s 3.3025 KOps/s $\color{#35bf28}+2.00\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0473ms 5.8017ms 172.3633 Ops/s 171.4891 Ops/s $\color{#35bf28}+0.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6911ms 0.3418ms 2.9256 KOps/s 3.1910 KOps/s $\textbf{\color{#d91a1a}-8.32\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5847ms 0.3261ms 3.0664 KOps/s 3.5035 KOps/s $\textbf{\color{#d91a1a}-12.48\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7189ms 1.4219ms 703.2886 Ops/s 741.7544 Ops/s $\textbf{\color{#d91a1a}-5.19\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.7719ms 1.3420ms 745.1782 Ops/s 795.1215 Ops/s $\textbf{\color{#d91a1a}-6.28\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.8710ms 6.0796ms 164.4833 Ops/s 167.8111 Ops/s $\color{#d91a1a}-1.98\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.0882ms 0.4537ms 2.2041 KOps/s 2.0951 KOps/s $\textbf{\color{#35bf28}+5.20\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8088ms 0.4482ms 2.2313 KOps/s 2.2815 KOps/s $\color{#d91a1a}-2.20\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0482ms 5.8485ms 170.9830 Ops/s 170.7332 Ops/s $\color{#35bf28}+0.15\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8861ms 0.3306ms 3.0247 KOps/s 2.7956 KOps/s $\textbf{\color{#35bf28}+8.20\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4719ms 0.2661ms 3.7577 KOps/s 2.6632 KOps/s $\textbf{\color{#35bf28}+41.10\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0484ms 5.8176ms 171.8925 Ops/s 171.8809 Ops/s $+0.01\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8564ms 0.2825ms 3.5400 KOps/s 3.5201 KOps/s $\color{#35bf28}+0.56\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4602ms 0.2659ms 3.7604 KOps/s 3.3375 KOps/s $\textbf{\color{#35bf28}+12.67\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 8.1015ms 5.9606ms 167.7671 Ops/s 165.8257 Ops/s $\color{#35bf28}+1.17\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8099ms 0.4986ms 2.0058 KOps/s 2.0552 KOps/s $\color{#d91a1a}-2.40\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8831ms 0.4832ms 2.0694 KOps/s 2.1326 KOps/s $\color{#d91a1a}-2.96\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5784s 16.5414ms 60.4545 Ops/s 198.8818 Ops/s $\textbf{\color{#d91a1a}-69.60\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 11.6319ms 2.0227ms 494.3896 Ops/s 534.6290 Ops/s $\textbf{\color{#d91a1a}-7.53\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.1236ms 1.0863ms 920.5404 Ops/s 767.4006 Ops/s $\textbf{\color{#35bf28}+19.96\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 8.5537ms 5.0716ms 197.1778 Ops/s 197.4967 Ops/s $\color{#d91a1a}-0.16\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 4.1198ms 1.7476ms 572.2258 Ops/s 532.8389 Ops/s $\textbf{\color{#35bf28}+7.39\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 12.1023ms 1.3034ms 767.2033 Ops/s 778.9230 Ops/s $\color{#d91a1a}-1.50\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5405s 16.0585ms 62.2722 Ops/s 58.8719 Ops/s $\textbf{\color{#35bf28}+5.78\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 4.0527ms 1.8920ms 528.5382 Ops/s 484.2359 Ops/s $\textbf{\color{#35bf28}+9.15\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.9497ms 1.0735ms 931.5715 Ops/s 922.7641 Ops/s $\color{#35bf28}+0.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.4066ms 34.8937ms 28.6585 Ops/s 28.2116 Ops/s $\color{#35bf28}+1.58\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.7110ms 18.1496ms 55.0975 Ops/s 54.9396 Ops/s $\color{#35bf28}+0.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 39.5077ms 35.8123ms 27.9233 Ops/s 27.9341 Ops/s $\color{#d91a1a}-0.04\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.9942ms 18.4642ms 54.1588 Ops/s 54.7012 Ops/s $\color{#d91a1a}-0.99\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 38.5793ms 37.7226ms 26.5093 Ops/s 26.4146 Ops/s $\color{#35bf28}+0.36\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.9965ms 19.8297ms 50.4294 Ops/s 50.8143 Ops/s $\color{#d91a1a}-0.76\%$

@vmoens vmoens merged commit d411bc7 into main Jan 19, 2026
106 of 109 checks passed
@vmoens vmoens deleted the fix-windows-optdepts-ci branch April 20, 2026 20:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

bug Something isn't working 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. Tests Incomplete or broken unit tests

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant