rl
rl copied to clipboard
[Feature] TensorDictMap Query module
:link: Helpful Links
:test_tube: See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2305
- :page_facing_up: Preview Python docs built from this PR
Note: Links to docs will display an error until the docs builds have been completed.
:x: 2 New Failures, 5 Unrelated Failures
As of commit f51a5677e1bd5e30d515bd1bb515c2be131b97cd with merge base 194a5ff127eba6fadd945dcf805ad7d004510777 ():
NEW FAILURES - The following jobs have failed:
- Build Windows Wheels / pytorch/rl / upload / wheel-py3_9-cpu (gh)
ERROR: failed to solve: python:3.12-slim: failed to resolve source metadata for docker.io/library/python:3.12-slim: failed to copy: httpReadSeeker: failed open: unexpected status code https://registry-1.docker.io/v2/library/python/manifests/sha256:af4e85f1cac90dd3771e47292ea7c8a9830abfabbe4faa5c53f158854c2e819d: 429 Too Many Requests - Server message: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit - Habitat Tests on Linux / tests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t 1eeee06f3fe9d0802ab60b0b80615587d538b813c3e729fe36e32d23d764408b /exec failed with exit code 134
BROKEN TRUNK - The following jobs failed but were present on the merge base:
👉 Rebase onto the `viable/strict` branch to avoid these failures
- Build Windows Wheels / pytorch/rl / upload / wheel-py3_9-cuda11_8 (gh) (trunk failure)
##[error]Unable to find any artifacts for the associated workflow - Build Windows Wheels / pytorch/rl / upload / wheel-py3_9-cuda12_1 (gh) (trunk failure)
##[error]Unable to find any artifacts for the associated workflow - Build Windows Wheels / pytorch/rl / upload / wheel-py3_9-cuda12_4 (gh) (trunk failure)
##[error]Unable to find any artifacts for the associated workflow - Unit-tests on Linux / tests-olddeps (3.8, 11.6) / linux-job (gh) (trunk failure)
test/test_modules.py::TestMultiAgent::test_multiagent_mlp_tdparams[True-batch0-6-True-True-3] - Unit-tests on Windows / unittests-cpu / windows-job (gh) (trunk failure)
test/test_collector.py::TestCompile::test_compiled_policy[device0-compile_policy2-MultiSyncDataCollector]
This comment was automatically generated by Dr. CI and updates every 15 minutes.
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests
Total Benchmarks: 143. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}5$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.4191s | 0.4177s | 2.3942 Ops/s | 2.4032 Ops/s | $\color{#d91a1a}-0.38\%$ |
| test_transformed | 0.7079s | 0.6178s | 1.6187 Ops/s | 1.6896 Ops/s | $\color{#d91a1a}-4.20\%$ |
| test_serial | 1.4456s | 1.3637s | 0.7333 Ops/s | 0.7455 Ops/s | $\color{#d91a1a}-1.63\%$ |
| test_parallel | 1.2444s | 1.2257s | 0.8158 Ops/s | 0.7934 Ops/s | $\color{#35bf28}+2.83\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1883ms | 28.9159μs | 34.5831 KOps/s | 35.0267 KOps/s | $\color{#d91a1a}-1.27\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 50.5750μs | 17.1503μs | 58.3079 KOps/s | 58.2549 KOps/s | $\color{#35bf28}+0.09\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 78.5080μs | 16.1312μs | 61.9916 KOps/s | 62.3369 KOps/s | $\color{#d91a1a}-0.55\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 34.2240μs | 9.4978μs | 105.2879 KOps/s | 103.9227 KOps/s | $\color{#35bf28}+1.31\%$ |
| test_step_mdp_speed[True-True-False-True-True] | 89.8290μs | 31.1109μs | 32.1430 KOps/s | 32.3700 KOps/s | $\color{#d91a1a}-0.70\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 48.0900μs | 19.2937μs | 51.8304 KOps/s | 51.5877 KOps/s | $\color{#35bf28}+0.47\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 77.0440μs | 18.2701μs | 54.7343 KOps/s | 55.3095 KOps/s | $\color{#d91a1a}-1.04\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 53.4200μs | 11.6085μs | 86.1437 KOps/s | 85.2604 KOps/s | $\color{#35bf28}+1.04\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 0.1021ms | 33.0982μs | 30.2132 KOps/s | 30.2038 KOps/s | $\color{#35bf28}+0.03\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 73.2570μs | 21.3771μs | 46.7790 KOps/s | 46.5421 KOps/s | $\color{#35bf28}+0.51\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 46.0870μs | 18.4377μs | 54.2367 KOps/s | 55.0247 KOps/s | $\color{#d91a1a}-1.43\%$ |
| test_step_mdp_speed[True-False-True-False-False] | 61.6060μs | 11.6706μs | 85.6857 KOps/s | 86.9954 KOps/s | $\color{#d91a1a}-1.51\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 94.7780μs | 34.8780μs | 28.6713 KOps/s | 28.0992 KOps/s | $\color{#35bf28}+2.04\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 84.8090μs | 23.2466μs | 43.0170 KOps/s | 41.4849 KOps/s | $\color{#35bf28}+3.69\%$ |
| test_step_mdp_speed[True-False-False-False-True] | 64.4610μs | 20.1732μs | 49.5706 KOps/s | 49.8264 KOps/s | $\color{#d91a1a}-0.51\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 56.5060μs | 13.6572μs | 73.2214 KOps/s | 72.9916 KOps/s | $\color{#35bf28}+0.31\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 82.7550μs | 33.5387μs | 29.8163 KOps/s | 30.2006 KOps/s | $\color{#d91a1a}-1.27\%$ |
| test_step_mdp_speed[False-True-True-True-False] | 60.8840μs | 21.3751μs | 46.7834 KOps/s | 46.1270 KOps/s | $\color{#35bf28}+1.42\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 83.2060μs | 21.4570μs | 46.6048 KOps/s | 46.3629 KOps/s | $\color{#35bf28}+0.52\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 2.5544ms | 13.4673μs | 74.2537 KOps/s | 73.3619 KOps/s | $\color{#35bf28}+1.22\%$ |
| test_step_mdp_speed[False-True-False-True-True] | 0.1124ms | 35.0271μs | 28.5493 KOps/s | 28.4061 KOps/s | $\color{#35bf28}+0.50\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 68.2380μs | 23.4499μs | 42.6441 KOps/s | 42.2731 KOps/s | $\color{#35bf28}+0.88\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 58.0690μs | 23.6797μs | 42.2302 KOps/s | 42.2011 KOps/s | $\color{#35bf28}+0.07\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 63.4290μs | 15.4093μs | 64.8957 KOps/s | 64.2536 KOps/s | $\color{#35bf28}+1.00\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 94.4730μs | 36.8953μs | 27.1038 KOps/s | 26.6693 KOps/s | $\color{#35bf28}+1.63\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 78.4010μs | 25.5159μs | 39.1912 KOps/s | 38.3718 KOps/s | $\color{#35bf28}+2.14\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 63.5800μs | 23.5917μs | 42.3878 KOps/s | 42.1686 KOps/s | $\color{#35bf28}+0.52\%$ |
| test_step_mdp_speed[False-False-True-False-False] | 57.4380μs | 15.2649μs | 65.5100 KOps/s | 64.4493 KOps/s | $\color{#35bf28}+1.65\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 96.9310μs | 38.4461μs | 26.0104 KOps/s | 25.5391 KOps/s | $\color{#35bf28}+1.85\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 67.9270μs | 27.1804μs | 36.7913 KOps/s | 35.6226 KOps/s | $\color{#35bf28}+3.28\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 87.1730μs | 25.2856μs | 39.5482 KOps/s | 39.9204 KOps/s | $\color{#d91a1a}-0.93\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 70.6330μs | 17.2294μs | 58.0405 KOps/s | 56.8699 KOps/s | $\color{#35bf28}+2.06\%$ |
| test_values[generalized_advantage_estimate-True-True] | 10.5439ms | 9.3524ms | 106.9249 Ops/s | 101.5088 Ops/s | $\textbf{\color{#35bf28}+5.34\%}$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 38.0089ms | 35.9280ms | 27.8334 Ops/s | 29.8030 Ops/s | $\textbf{\color{#d91a1a}-6.61\%}$ |
| test_values[td0_return_estimate-False-False] | 0.2222ms | 0.1712ms | 5.8401 KOps/s | 5.5989 KOps/s | $\color{#35bf28}+4.31\%$ |
| test_values[td1_return_estimate-False-False] | 27.5913ms | 23.5383ms | 42.4839 Ops/s | 41.0513 Ops/s | $\color{#35bf28}+3.49\%$ |
| test_values[vec_td1_return_estimate-False-False] | 38.8905ms | 36.1354ms | 27.6737 Ops/s | 29.7987 Ops/s | $\textbf{\color{#d91a1a}-7.13\%}$ |
| test_values[td_lambda_return_estimate-True-False] | 35.2889ms | 33.8422ms | 29.5489 Ops/s | 28.7389 Ops/s | $\color{#35bf28}+2.82\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 39.0478ms | 36.1098ms | 27.6933 Ops/s | 29.7636 Ops/s | $\textbf{\color{#d91a1a}-6.96\%}$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 12.3070ms | 8.2566ms | 121.1149 Ops/s | 116.5545 Ops/s | $\color{#35bf28}+3.91\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.1042ms | 1.7773ms | 562.6659 Ops/s | 493.1469 Ops/s | $\textbf{\color{#35bf28}+14.10\%}$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.5150ms | 0.3534ms | 2.8296 KOps/s | 2.7921 KOps/s | $\color{#35bf28}+1.34\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 48.7545ms | 47.0107ms | 21.2717 Ops/s | 24.8745 Ops/s | $\textbf{\color{#d91a1a}-14.48\%}$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 3.7470ms | 3.0398ms | 328.9678 Ops/s | 326.6142 Ops/s | $\color{#35bf28}+0.72\%$ |
| test_dqn_speed[False-None] | 1.7913ms | 1.3422ms | 745.0731 Ops/s | 749.4554 Ops/s | $\color{#d91a1a}-0.58\%$ |
| test_dqn_speed[False-backward] | 2.1509ms | 1.8410ms | 543.1975 Ops/s | 545.8213 Ops/s | $\color{#d91a1a}-0.48\%$ |
| test_dqn_speed[True-None] | 1.2446ms | 0.4663ms | 2.1446 KOps/s | 2.0684 KOps/s | $\color{#35bf28}+3.68\%$ |
| test_dqn_speed[True-backward] | 0.9224ms | 0.8818ms | 1.1341 KOps/s | 1.1121 KOps/s | $\color{#35bf28}+1.98\%$ |
| test_dqn_speed[reduce-overhead-None] | 0.6845ms | 0.4698ms | 2.1286 KOps/s | 2.1372 KOps/s | $\color{#d91a1a}-0.41\%$ |
| test_dqn_speed[reduce-overhead-backward] | 0.9355ms | 0.8772ms | 1.1400 KOps/s | 1.1004 KOps/s | $\color{#35bf28}+3.60\%$ |
| test_ddpg_speed[False-None] | 3.8384ms | 2.7898ms | 358.4441 Ops/s | 352.3652 Ops/s | $\color{#35bf28}+1.73\%$ |
| test_ddpg_speed[False-backward] | 4.3303ms | 3.9496ms | 253.1882 Ops/s | 241.8656 Ops/s | $\color{#35bf28}+4.68\%$ |
| test_ddpg_speed[True-None] | 1.3577ms | 1.0220ms | 978.4809 Ops/s | 983.4312 Ops/s | $\color{#d91a1a}-0.50\%$ |
| test_ddpg_speed[True-backward] | 2.0288ms | 1.9278ms | 518.7141 Ops/s | 523.5745 Ops/s | $\color{#d91a1a}-0.93\%$ |
| test_ddpg_speed[reduce-overhead-None] | 1.3107ms | 1.0247ms | 975.8769 Ops/s | 987.2365 Ops/s | $\color{#d91a1a}-1.15\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.3907ms | 1.9660ms | 508.6502 Ops/s | 523.0039 Ops/s | $\color{#d91a1a}-2.74\%$ |
| test_sac_speed[False-None] | 9.6999ms | 8.0578ms | 124.1041 Ops/s | 123.5267 Ops/s | $\color{#35bf28}+0.47\%$ |
| test_sac_speed[False-backward] | 11.2675ms | 10.8078ms | 92.5260 Ops/s | 91.6661 Ops/s | $\color{#35bf28}+0.94\%$ |
| test_sac_speed[True-None] | 2.4805ms | 1.8675ms | 535.4863 Ops/s | 535.0530 Ops/s | $\color{#35bf28}+0.08\%$ |
| test_sac_speed[True-backward] | 3.6815ms | 3.5719ms | 279.9636 Ops/s | 279.3726 Ops/s | $\color{#35bf28}+0.21\%$ |
| test_sac_speed[reduce-overhead-None] | 5.8541ms | 1.8885ms | 529.5261 Ops/s | 533.1266 Ops/s | $\color{#d91a1a}-0.68\%$ |
| test_sac_speed[reduce-overhead-backward] | 3.7058ms | 3.6053ms | 277.3668 Ops/s | 275.0708 Ops/s | $\color{#35bf28}+0.83\%$ |
| test_redq_speed[False-None] | 18.1836ms | 13.2197ms | 75.6446 Ops/s | 76.2478 Ops/s | $\color{#d91a1a}-0.79\%$ |
| test_redq_speed[False-backward] | 27.1215ms | 22.5959ms | 44.2558 Ops/s | 44.3158 Ops/s | $\color{#d91a1a}-0.14\%$ |
| test_redq_speed[True-None] | 5.6742ms | 4.9620ms | 201.5332 Ops/s | 204.6146 Ops/s | $\color{#d91a1a}-1.51\%$ |
| test_redq_speed[True-backward] | 13.4955ms | 12.5289ms | 79.8156 Ops/s | 80.4837 Ops/s | $\color{#d91a1a}-0.83\%$ |
| test_redq_speed[reduce-overhead-None] | 6.0316ms | 5.1216ms | 195.2512 Ops/s | 204.9644 Ops/s | $\color{#d91a1a}-4.74\%$ |
| test_redq_speed[reduce-overhead-backward] | 13.3221ms | 12.6282ms | 79.1879 Ops/s | 80.5231 Ops/s | $\color{#d91a1a}-1.66\%$ |
| test_redq_deprec_speed[False-None] | 15.1300ms | 12.9355ms | 77.3069 Ops/s | 77.7242 Ops/s | $\color{#d91a1a}-0.54\%$ |
| test_redq_deprec_speed[False-backward] | 20.2482ms | 18.8572ms | 53.0302 Ops/s | 53.3842 Ops/s | $\color{#d91a1a}-0.66\%$ |
| test_redq_deprec_speed[True-None] | 4.2981ms | 3.7361ms | 267.6586 Ops/s | 269.6950 Ops/s | $\color{#d91a1a}-0.76\%$ |
| test_redq_deprec_speed[True-backward] | 9.4172ms | 8.6326ms | 115.8406 Ops/s | 120.0898 Ops/s | $\color{#d91a1a}-3.54\%$ |
| test_redq_deprec_speed[reduce-overhead-None] | 4.3492ms | 3.6738ms | 272.1940 Ops/s | 275.5025 Ops/s | $\color{#d91a1a}-1.20\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 9.6369ms | 8.5491ms | 116.9715 Ops/s | 122.3062 Ops/s | $\color{#d91a1a}-4.36\%$ |
| test_td3_speed[False-None] | 9.1259ms | 7.8525ms | 127.3488 Ops/s | 124.9256 Ops/s | $\color{#35bf28}+1.94\%$ |
| test_td3_speed[False-backward] | 10.6214ms | 10.1665ms | 98.3620 Ops/s | 96.1021 Ops/s | $\color{#35bf28}+2.35\%$ |
| test_td3_speed[True-None] | 1.9067ms | 1.7616ms | 567.6815 Ops/s | 560.6387 Ops/s | $\color{#35bf28}+1.26\%$ |
| test_td3_speed[True-backward] | 3.7942ms | 3.4062ms | 293.5843 Ops/s | 285.0962 Ops/s | $\color{#35bf28}+2.98\%$ |
| test_td3_speed[reduce-overhead-None] | 2.1031ms | 1.7711ms | 564.6246 Ops/s | 564.4405 Ops/s | $\color{#35bf28}+0.03\%$ |
| test_td3_speed[reduce-overhead-backward] | 5.0346ms | 3.4323ms | 291.3471 Ops/s | 295.3158 Ops/s | $\color{#d91a1a}-1.34\%$ |
| test_cql_speed[False-None] | 55.9261ms | 37.0979ms | 26.9557 Ops/s | 27.6836 Ops/s | $\color{#d91a1a}-2.63\%$ |
| test_cql_speed[False-backward] | 51.8645ms | 46.8956ms | 21.3240 Ops/s | 21.1230 Ops/s | $\color{#35bf28}+0.95\%$ |
| test_cql_speed[True-None] | 18.3753ms | 15.8985ms | 62.8992 Ops/s | 62.8656 Ops/s | $\color{#35bf28}+0.05\%$ |
| test_cql_speed[True-backward] | 29.5467ms | 22.7415ms | 43.9725 Ops/s | 44.4570 Ops/s | $\color{#d91a1a}-1.09\%$ |
| test_cql_speed[reduce-overhead-None] | 16.5070ms | 15.9875ms | 62.5487 Ops/s | 63.0601 Ops/s | $\color{#d91a1a}-0.81\%$ |
| test_cql_speed[reduce-overhead-backward] | 23.7902ms | 22.5582ms | 44.3297 Ops/s | 44.1175 Ops/s | $\color{#35bf28}+0.48\%$ |
| test_a2c_speed[False-None] | 8.7233ms | 7.2422ms | 138.0801 Ops/s | 134.6263 Ops/s | $\color{#35bf28}+2.57\%$ |
| test_a2c_speed[False-backward] | 14.6714ms | 14.4037ms | 69.4266 Ops/s | 67.5765 Ops/s | $\color{#35bf28}+2.74\%$ |
| test_a2c_speed[True-None] | 3.7674ms | 3.3480ms | 298.6836 Ops/s | 294.3162 Ops/s | $\color{#35bf28}+1.48\%$ |
| test_a2c_speed[True-backward] | 10.4115ms | 9.9330ms | 100.6742 Ops/s | 98.2914 Ops/s | $\color{#35bf28}+2.42\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.6288ms | 3.3221ms | 301.0101 Ops/s | 296.7163 Ops/s | $\color{#35bf28}+1.45\%$ |
| test_a2c_speed[reduce-overhead-backward] | 10.7815ms | 10.0891ms | 99.1172 Ops/s | 99.5740 Ops/s | $\color{#d91a1a}-0.46\%$ |
| test_ppo_speed[False-None] | 9.8621ms | 7.5513ms | 132.4278 Ops/s | 130.7668 Ops/s | $\color{#35bf28}+1.27\%$ |
| test_ppo_speed[False-backward] | 15.6301ms | 15.0313ms | 66.5277 Ops/s | 65.1749 Ops/s | $\color{#35bf28}+2.08\%$ |
| test_ppo_speed[True-None] | 4.0817ms | 3.7690ms | 265.3219 Ops/s | 265.9848 Ops/s | $\color{#d91a1a}-0.25\%$ |
| test_ppo_speed[True-backward] | 12.2631ms | 10.1659ms | 98.3678 Ops/s | 101.7844 Ops/s | $\color{#d91a1a}-3.36\%$ |
| test_ppo_speed[reduce-overhead-None] | 3.9961ms | 3.7405ms | 267.3409 Ops/s | 264.2850 Ops/s | $\color{#35bf28}+1.16\%$ |
| test_ppo_speed[reduce-overhead-backward] | 10.9868ms | 10.0577ms | 99.4263 Ops/s | 97.1241 Ops/s | $\color{#35bf28}+2.37\%$ |
| test_reinforce_speed[False-None] | 7.2190ms | 6.4994ms | 153.8610 Ops/s | 149.4317 Ops/s | $\color{#35bf28}+2.96\%$ |
| test_reinforce_speed[False-backward] | 11.8644ms | 9.8653ms | 101.3654 Ops/s | 99.8862 Ops/s | $\color{#35bf28}+1.48\%$ |
| test_reinforce_speed[True-None] | 3.1640ms | 2.7484ms | 363.8430 Ops/s | 367.4071 Ops/s | $\color{#d91a1a}-0.97\%$ |
| test_reinforce_speed[True-backward] | 9.5238ms | 8.9784ms | 111.3786 Ops/s | 109.1430 Ops/s | $\color{#35bf28}+2.05\%$ |
| test_reinforce_speed[reduce-overhead-None] | 3.5115ms | 2.7662ms | 361.5075 Ops/s | 367.7668 Ops/s | $\color{#d91a1a}-1.70\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 9.5677ms | 9.0269ms | 110.7799 Ops/s | 110.6119 Ops/s | $\color{#35bf28}+0.15\%$ |
| test_iql_speed[False-None] | 33.9317ms | 32.4360ms | 30.8299 Ops/s | 29.9723 Ops/s | $\color{#35bf28}+2.86\%$ |
| test_iql_speed[False-backward] | 45.8301ms | 44.5059ms | 22.4689 Ops/s | 21.8812 Ops/s | $\color{#35bf28}+2.69\%$ |
| test_iql_speed[True-None] | 14.1747ms | 13.6241ms | 73.3995 Ops/s | 72.6267 Ops/s | $\color{#35bf28}+1.06\%$ |
| test_iql_speed[True-backward] | 26.3043ms | 25.1693ms | 39.7309 Ops/s | 39.8868 Ops/s | $\color{#d91a1a}-0.39\%$ |
| test_iql_speed[reduce-overhead-None] | 14.3371ms | 13.8050ms | 72.4375 Ops/s | 72.0649 Ops/s | $\color{#35bf28}+0.52\%$ |
| test_iql_speed[reduce-overhead-backward] | 30.4845ms | 25.7728ms | 38.8005 Ops/s | 39.5900 Ops/s | $\color{#d91a1a}-1.99\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 5.3117ms | 4.9357ms | 202.6036 Ops/s | 198.3574 Ops/s | $\color{#35bf28}+2.14\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.9786ms | 0.4904ms | 2.0391 KOps/s | 2.0428 KOps/s | $\color{#d91a1a}-0.18\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6931ms | 0.4694ms | 2.1305 KOps/s | 2.1467 KOps/s | $\color{#d91a1a}-0.75\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.9317ms | 4.9451ms | 202.2187 Ops/s | 197.4770 Ops/s | $\color{#35bf28}+2.40\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7858ms | 0.4835ms | 2.0681 KOps/s | 2.0571 KOps/s | $\color{#35bf28}+0.54\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7034ms | 0.4625ms | 2.1621 KOps/s | 2.1547 KOps/s | $\color{#35bf28}+0.34\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 2.0370ms | 1.5773ms | 633.9755 Ops/s | 622.9575 Ops/s | $\color{#35bf28}+1.77\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.8266ms | 1.5269ms | 654.9367 Ops/s | 642.2732 Ops/s | $\color{#35bf28}+1.97\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 7.0141ms | 5.0991ms | 196.1117 Ops/s | 190.3455 Ops/s | $\color{#35bf28}+3.03\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.5568ms | 0.6206ms | 1.6113 KOps/s | 1.5872 KOps/s | $\color{#35bf28}+1.51\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8488ms | 0.5987ms | 1.6704 KOps/s | 1.6628 KOps/s | $\color{#35bf28}+0.45\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 5.3972ms | 4.8671ms | 205.4604 Ops/s | 193.0289 Ops/s | $\textbf{\color{#35bf28}+6.44\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7067ms | 0.4870ms | 2.0532 KOps/s | 2.0483 KOps/s | $\color{#35bf28}+0.24\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 7.8793ms | 0.4766ms | 2.0981 KOps/s | 2.0987 KOps/s | $\color{#d91a1a}-0.03\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 7.7125ms | 4.9822ms | 200.7150 Ops/s | 189.5772 Ops/s | $\textbf{\color{#35bf28}+5.88\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.5171ms | 0.4903ms | 2.0394 KOps/s | 2.0410 KOps/s | $\color{#d91a1a}-0.08\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6758ms | 0.4719ms | 2.1191 KOps/s | 2.1809 KOps/s | $\color{#d91a1a}-2.83\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.4286ms | 5.2222ms | 191.4920 Ops/s | 192.3443 Ops/s | $\color{#d91a1a}-0.44\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.4461ms | 0.6352ms | 1.5744 KOps/s | 1.5746 KOps/s | $\color{#d91a1a}-0.01\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8338ms | 0.6053ms | 1.6519 KOps/s | 1.6488 KOps/s | $\color{#35bf28}+0.19\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4770s | 13.7694ms | 72.6246 Ops/s | 220.4532 Ops/s | $\textbf{\color{#d91a1a}-67.06\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 9.2699ms | 2.3433ms | 426.7425 Ops/s | 392.7470 Ops/s | $\textbf{\color{#35bf28}+8.66\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 1.8772ms | 1.1927ms | 838.4455 Ops/s | 786.9968 Ops/s | $\textbf{\color{#35bf28}+6.54\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 6.1469ms | 4.2650ms | 234.4655 Ops/s | 233.2310 Ops/s | $\color{#35bf28}+0.53\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 7.0432ms | 2.3522ms | 425.1423 Ops/s | 432.5034 Ops/s | $\color{#d91a1a}-1.70\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 7.2864ms | 1.3280ms | 753.0207 Ops/s | 772.2336 Ops/s | $\color{#d91a1a}-2.49\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.3954s | 12.4456ms | 80.3496 Ops/s | 33.8948 Ops/s | $\textbf{\color{#35bf28}+137.06\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 7.9909ms | 2.4626ms | 406.0733 Ops/s | 424.6032 Ops/s | $\color{#d91a1a}-4.36\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.7606ms | 1.4515ms | 688.9637 Ops/s | 659.5056 Ops/s | $\color{#35bf28}+4.47\%$ |
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests
Total Benchmarks: 143. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}13$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.7336s | 0.7231s | 1.3830 Ops/s | 1.4008 Ops/s | $\color{#d91a1a}-1.27\%$ |
| test_transformed | 1.0407s | 0.9673s | 1.0338 Ops/s | 1.0570 Ops/s | $\color{#d91a1a}-2.20\%$ |
| test_serial | 2.1481s | 2.0717s | 0.4827 Ops/s | 0.4842 Ops/s | $\color{#d91a1a}-0.30\%$ |
| test_parallel | 1.9422s | 1.8684s | 0.5352 Ops/s | 0.5377 Ops/s | $\color{#d91a1a}-0.47\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1949ms | 37.9364μs | 26.3599 KOps/s | 26.6075 KOps/s | $\color{#d91a1a}-0.93\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 0.2151ms | 22.3081μs | 44.8268 KOps/s | 45.1059 KOps/s | $\color{#d91a1a}-0.62\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 47.9710μs | 20.3157μs | 49.2231 KOps/s | 50.3317 KOps/s | $\color{#d91a1a}-2.20\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 65.7510μs | 11.6162μs | 86.0865 KOps/s | 84.9230 KOps/s | $\color{#35bf28}+1.37\%$ |
| test_step_mdp_speed[True-True-False-True-True] | 0.1062ms | 40.4994μs | 24.6917 KOps/s | 25.2890 KOps/s | $\color{#d91a1a}-2.36\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 45.7600μs | 24.4440μs | 40.9098 KOps/s | 40.9879 KOps/s | $\color{#d91a1a}-0.19\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 60.3410μs | 23.3670μs | 42.7955 KOps/s | 43.8683 KOps/s | $\color{#d91a1a}-2.45\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 48.9110μs | 14.4236μs | 69.3307 KOps/s | 69.1266 KOps/s | $\color{#35bf28}+0.30\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 0.1725ms | 42.4046μs | 23.5823 KOps/s | 23.3635 KOps/s | $\color{#35bf28}+0.94\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 62.2010μs | 27.1129μs | 36.8828 KOps/s | 36.7806 KOps/s | $\color{#35bf28}+0.28\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 85.1010μs | 22.8465μs | 43.7704 KOps/s | 44.9467 KOps/s | $\color{#d91a1a}-2.62\%$ |
| test_step_mdp_speed[True-False-True-False-False] | 0.4194ms | 14.3097μs | 69.8825 KOps/s | 69.8389 KOps/s | $\color{#35bf28}+0.06\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 88.7910μs | 45.2239μs | 22.1122 KOps/s | 22.3474 KOps/s | $\color{#d91a1a}-1.05\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 0.1138ms | 29.5953μs | 33.7891 KOps/s | 34.1455 KOps/s | $\color{#d91a1a}-1.04\%$ |
| test_step_mdp_speed[True-False-False-False-True] | 52.5310μs | 25.5103μs | 39.1999 KOps/s | 40.1369 KOps/s | $\color{#d91a1a}-2.33\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 44.2200μs | 17.1208μs | 58.4084 KOps/s | 58.2005 KOps/s | $\color{#35bf28}+0.36\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 82.5010μs | 42.8858μs | 23.3177 KOps/s | 23.4674 KOps/s | $\color{#d91a1a}-0.64\%$ |
| test_step_mdp_speed[False-True-True-True-False] | 89.3910μs | 26.6350μs | 37.5446 KOps/s | 36.9582 KOps/s | $\color{#35bf28}+1.59\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 57.4710μs | 26.9729μs | 37.0743 KOps/s | 36.3877 KOps/s | $\color{#35bf28}+1.89\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 2.9506ms | 17.1863μs | 58.1857 KOps/s | 51.6055 KOps/s | $\textbf{\color{#35bf28}+12.75\%}$ |
| test_step_mdp_speed[False-True-False-True-True] | 0.1726ms | 46.0096μs | 21.7346 KOps/s | 21.9431 KOps/s | $\color{#d91a1a}-0.95\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 55.0810μs | 29.4761μs | 33.9258 KOps/s | 33.7582 KOps/s | $\color{#35bf28}+0.50\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 64.7010μs | 30.7166μs | 32.5557 KOps/s | 33.3576 KOps/s | $\color{#d91a1a}-2.40\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 41.8310μs | 19.5805μs | 51.0713 KOps/s | 52.2297 KOps/s | $\color{#d91a1a}-2.22\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 79.8220μs | 48.4908μs | 20.6225 KOps/s | 21.1114 KOps/s | $\color{#d91a1a}-2.32\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 63.7110μs | 31.8580μs | 31.3893 KOps/s | 31.1929 KOps/s | $\color{#35bf28}+0.63\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 56.5310μs | 29.5400μs | 33.8524 KOps/s | 34.4793 KOps/s | $\color{#d91a1a}-1.82\%$ |
| test_step_mdp_speed[False-False-True-False-False] | 0.1733ms | 19.5015μs | 51.2782 KOps/s | 52.4827 KOps/s | $\color{#d91a1a}-2.30\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 99.2820μs | 50.5027μs | 19.8009 KOps/s | 20.1347 KOps/s | $\color{#d91a1a}-1.66\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 63.0810μs | 34.9034μs | 28.6505 KOps/s | 29.0350 KOps/s | $\color{#d91a1a}-1.32\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 65.6710μs | 32.9251μs | 30.3720 KOps/s | 31.2552 KOps/s | $\color{#d91a1a}-2.83\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 56.4510μs | 21.9408μs | 45.5773 KOps/s | 46.2604 KOps/s | $\color{#d91a1a}-1.48\%$ |
| test_values[generalized_advantage_estimate-True-True] | 25.0638ms | 24.6704ms | 40.5344 Ops/s | 40.0719 Ops/s | $\color{#35bf28}+1.15\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 0.1023s | 2.9267ms | 341.6830 Ops/s | 360.2288 Ops/s | $\textbf{\color{#d91a1a}-5.15\%}$ |
| test_values[td0_return_estimate-False-False] | 84.8210μs | 64.6536μs | 15.4670 KOps/s | 15.0390 KOps/s | $\color{#35bf28}+2.85\%$ |
| test_values[td1_return_estimate-False-False] | 55.1343ms | 54.4572ms | 18.3630 Ops/s | 17.6491 Ops/s | $\color{#35bf28}+4.05\%$ |
| test_values[vec_td1_return_estimate-False-False] | 1.4118ms | 1.0639ms | 939.9547 Ops/s | 938.7133 Ops/s | $\color{#35bf28}+0.13\%$ |
| test_values[td_lambda_return_estimate-True-False] | 87.1889ms | 86.1652ms | 11.6056 Ops/s | 11.6596 Ops/s | $\color{#d91a1a}-0.46\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 1.3865ms | 1.0580ms | 945.1375 Ops/s | 938.7483 Ops/s | $\color{#35bf28}+0.68\%$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.7180ms | 24.3421ms | 41.0811 Ops/s | 41.1595 Ops/s | $\color{#d91a1a}-0.19\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.0512ms | 0.7291ms | 1.3715 KOps/s | 1.3731 KOps/s | $\color{#d91a1a}-0.12\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.8033ms | 0.6457ms | 1.5487 KOps/s | 1.5494 KOps/s | $\color{#d91a1a}-0.05\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6053ms | 1.4602ms | 684.8348 Ops/s | 686.4341 Ops/s | $\color{#d91a1a}-0.23\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.8318ms | 0.6846ms | 1.4607 KOps/s | 1.5086 KOps/s | $\color{#d91a1a}-3.18\%$ |
| test_dqn_speed[False-None] | 7.1599ms | 1.3175ms | 759.0046 Ops/s | 757.6091 Ops/s | $\color{#35bf28}+0.18\%$ |
| test_dqn_speed[False-backward] | 2.0296ms | 1.8360ms | 544.6754 Ops/s | 538.0500 Ops/s | $\color{#35bf28}+1.23\%$ |
| test_dqn_speed[True-None] | 0.7180ms | 0.5614ms | 1.7813 KOps/s | 1.7804 KOps/s | $\color{#35bf28}+0.05\%$ |
| test_dqn_speed[True-backward] | 1.0560ms | 1.0068ms | 993.2889 Ops/s | 873.5478 Ops/s | $\textbf{\color{#35bf28}+13.71\%}$ |
| test_dqn_speed[reduce-overhead-None] | 0.7106ms | 0.5571ms | 1.7949 KOps/s | 1.7302 KOps/s | $\color{#35bf28}+3.74\%$ |
| test_dqn_speed[reduce-overhead-backward] | 1.1100ms | 1.0233ms | 977.2509 Ops/s | 976.2229 Ops/s | $\color{#35bf28}+0.11\%$ |
| test_ddpg_speed[False-None] | 3.3936ms | 2.6857ms | 372.3428 Ops/s | 369.3441 Ops/s | $\color{#35bf28}+0.81\%$ |
| test_ddpg_speed[False-backward] | 4.2808ms | 3.9929ms | 250.4449 Ops/s | 252.0412 Ops/s | $\color{#d91a1a}-0.63\%$ |
| test_ddpg_speed[True-None] | 1.5681ms | 1.2613ms | 792.8620 Ops/s | 790.3255 Ops/s | $\color{#35bf28}+0.32\%$ |
| test_ddpg_speed[True-backward] | 2.4462ms | 2.2898ms | 436.7129 Ops/s | 444.2785 Ops/s | $\color{#d91a1a}-1.70\%$ |
| test_ddpg_speed[reduce-overhead-None] | 1.4802ms | 1.2722ms | 786.0662 Ops/s | 780.9594 Ops/s | $\color{#35bf28}+0.65\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.4569ms | 2.2687ms | 440.7879 Ops/s | 444.4018 Ops/s | $\color{#d91a1a}-0.81\%$ |
| test_sac_speed[False-None] | 7.8737ms | 7.5395ms | 132.6340 Ops/s | 129.9235 Ops/s | $\color{#35bf28}+2.09\%$ |
| test_sac_speed[False-backward] | 11.3015ms | 10.8818ms | 91.8963 Ops/s | 91.2623 Ops/s | $\color{#35bf28}+0.69\%$ |
| test_sac_speed[True-None] | 2.2468ms | 2.0673ms | 483.7281 Ops/s | 482.2828 Ops/s | $\color{#35bf28}+0.30\%$ |
| test_sac_speed[True-backward] | 4.4621ms | 4.0482ms | 247.0227 Ops/s | 235.6173 Ops/s | $\color{#35bf28}+4.84\%$ |
| test_sac_speed[reduce-overhead-None] | 2.2453ms | 2.0701ms | 483.0609 Ops/s | 481.0060 Ops/s | $\color{#35bf28}+0.43\%$ |
| test_sac_speed[reduce-overhead-backward] | 4.2901ms | 4.0774ms | 245.2547 Ops/s | 244.7340 Ops/s | $\color{#35bf28}+0.21\%$ |
| test_redq_speed[False-None] | 15.8306ms | 10.8579ms | 92.0986 Ops/s | 83.0241 Ops/s | $\textbf{\color{#35bf28}+10.93\%}$ |
| test_redq_speed[False-backward] | 19.2665ms | 17.8094ms | 56.1501 Ops/s | 54.9731 Ops/s | $\color{#35bf28}+2.14\%$ |
| test_redq_speed[True-None] | 3.9002ms | 3.7000ms | 270.2733 Ops/s | 268.5700 Ops/s | $\color{#35bf28}+0.63\%$ |
| test_redq_speed[True-backward] | 9.4550ms | 8.8789ms | 112.6271 Ops/s | 112.8067 Ops/s | $\color{#d91a1a}-0.16\%$ |
| test_redq_speed[reduce-overhead-None] | 4.0074ms | 3.6780ms | 271.8865 Ops/s | 275.0147 Ops/s | $\color{#d91a1a}-1.14\%$ |
| test_redq_speed[reduce-overhead-backward] | 9.2347ms | 8.8359ms | 113.1742 Ops/s | 108.5747 Ops/s | $\color{#35bf28}+4.24\%$ |
| test_redq_deprec_speed[False-None] | 11.3506ms | 10.6577ms | 93.8290 Ops/s | 91.8625 Ops/s | $\color{#35bf28}+2.14\%$ |
| test_redq_deprec_speed[False-backward] | 16.3179ms | 15.7101ms | 63.6531 Ops/s | 63.2897 Ops/s | $\color{#35bf28}+0.57\%$ |
| test_redq_deprec_speed[True-None] | 3.6706ms | 3.2787ms | 304.9969 Ops/s | 296.2755 Ops/s | $\color{#35bf28}+2.94\%$ |
| test_redq_deprec_speed[True-backward] | 7.4907ms | 7.2526ms | 137.8809 Ops/s | 138.2509 Ops/s | $\color{#d91a1a}-0.27\%$ |
| test_redq_deprec_speed[reduce-overhead-None] | 3.5050ms | 3.2447ms | 308.1910 Ops/s | 310.2653 Ops/s | $\color{#d91a1a}-0.67\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 7.5454ms | 7.2297ms | 138.3185 Ops/s | 143.8504 Ops/s | $\color{#d91a1a}-3.85\%$ |
| test_td3_speed[False-None] | 8.3148ms | 7.4944ms | 133.4335 Ops/s | 131.2021 Ops/s | $\color{#35bf28}+1.70\%$ |
| test_td3_speed[False-backward] | 10.7944ms | 10.3692ms | 96.4393 Ops/s | 95.0517 Ops/s | $\color{#35bf28}+1.46\%$ |
| test_td3_speed[True-None] | 2.0091ms | 1.9512ms | 512.5056 Ops/s | 508.1797 Ops/s | $\color{#35bf28}+0.85\%$ |
| test_td3_speed[True-backward] | 3.9911ms | 3.8232ms | 261.5620 Ops/s | 225.6041 Ops/s | $\textbf{\color{#35bf28}+15.94\%}$ |
| test_td3_speed[reduce-overhead-None] | 2.0039ms | 1.9564ms | 511.1346 Ops/s | 507.2911 Ops/s | $\color{#35bf28}+0.76\%$ |
| test_td3_speed[reduce-overhead-backward] | 4.0031ms | 3.8084ms | 262.5763 Ops/s | 259.7502 Ops/s | $\color{#35bf28}+1.09\%$ |
| test_cql_speed[False-None] | 27.7422ms | 25.3957ms | 39.3768 Ops/s | 39.3292 Ops/s | $\color{#35bf28}+0.12\%$ |
| test_cql_speed[False-backward] | 39.1274ms | 35.2065ms | 28.4038 Ops/s | 28.7373 Ops/s | $\color{#d91a1a}-1.16\%$ |
| test_cql_speed[True-None] | 11.6562ms | 11.1831ms | 89.4208 Ops/s | 88.3909 Ops/s | $\color{#35bf28}+1.17\%$ |
| test_cql_speed[True-backward] | 18.0512ms | 17.1912ms | 58.1692 Ops/s | 56.7822 Ops/s | $\color{#35bf28}+2.44\%$ |
| test_cql_speed[reduce-overhead-None] | 11.8435ms | 11.3511ms | 88.0973 Ops/s | 88.7065 Ops/s | $\color{#d91a1a}-0.69\%$ |
| test_cql_speed[reduce-overhead-backward] | 17.8820ms | 17.1623ms | 58.2673 Ops/s | 57.2412 Ops/s | $\color{#35bf28}+1.79\%$ |
| test_a2c_speed[False-None] | 7.5581ms | 5.4185ms | 184.5537 Ops/s | 180.4189 Ops/s | $\color{#35bf28}+2.29\%$ |
| test_a2c_speed[False-backward] | 12.3166ms | 12.0908ms | 82.7073 Ops/s | 82.4229 Ops/s | $\color{#35bf28}+0.35\%$ |
| test_a2c_speed[True-None] | 3.4475ms | 3.1647ms | 315.9822 Ops/s | 314.6107 Ops/s | $\color{#35bf28}+0.44\%$ |
| test_a2c_speed[True-backward] | 8.9123ms | 8.6982ms | 114.9667 Ops/s | 113.7288 Ops/s | $\color{#35bf28}+1.09\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.5087ms | 3.1183ms | 320.6884 Ops/s | 321.1352 Ops/s | $\color{#d91a1a}-0.14\%$ |
| test_a2c_speed[reduce-overhead-backward] | 9.0240ms | 8.7106ms | 114.8025 Ops/s | 114.8653 Ops/s | $\color{#d91a1a}-0.05\%$ |
| test_ppo_speed[False-None] | 6.0156ms | 5.7621ms | 173.5480 Ops/s | 168.3630 Ops/s | $\color{#35bf28}+3.08\%$ |
| test_ppo_speed[False-backward] | 12.9948ms | 12.5691ms | 79.5603 Ops/s | 77.5435 Ops/s | $\color{#35bf28}+2.60\%$ |
| test_ppo_speed[True-None] | 3.6773ms | 3.4910ms | 286.4478 Ops/s | 283.1978 Ops/s | $\color{#35bf28}+1.15\%$ |
| test_ppo_speed[True-backward] | 8.5925ms | 8.3705ms | 119.4678 Ops/s | 119.0137 Ops/s | $\color{#35bf28}+0.38\%$ |
| test_ppo_speed[reduce-overhead-None] | 3.6886ms | 3.4994ms | 285.7607 Ops/s | 283.0152 Ops/s | $\color{#35bf28}+0.97\%$ |
| test_ppo_speed[reduce-overhead-backward] | 8.4994ms | 8.3224ms | 120.1569 Ops/s | 119.2447 Ops/s | $\color{#35bf28}+0.77\%$ |
| test_reinforce_speed[False-None] | 6.4803ms | 4.5084ms | 221.8087 Ops/s | 217.7572 Ops/s | $\color{#35bf28}+1.86\%$ |
| test_reinforce_speed[False-backward] | 7.7442ms | 7.3649ms | 135.7799 Ops/s | 135.0992 Ops/s | $\color{#35bf28}+0.50\%$ |
| test_reinforce_speed[True-None] | 2.5021ms | 2.2809ms | 438.4248 Ops/s | 444.1627 Ops/s | $\color{#d91a1a}-1.29\%$ |
| test_reinforce_speed[True-backward] | 7.5105ms | 7.2701ms | 137.5496 Ops/s | 135.7723 Ops/s | $\color{#35bf28}+1.31\%$ |
| test_reinforce_speed[reduce-overhead-None] | 2.5012ms | 2.2909ms | 436.5015 Ops/s | 435.4280 Ops/s | $\color{#35bf28}+0.25\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 7.3828ms | 7.1713ms | 139.4450 Ops/s | 137.6866 Ops/s | $\color{#35bf28}+1.28\%$ |
| test_iql_speed[False-None] | 25.7008ms | 20.1697ms | 49.5794 Ops/s | 49.2946 Ops/s | $\color{#35bf28}+0.58\%$ |
| test_iql_speed[False-backward] | 36.1839ms | 30.7447ms | 32.5259 Ops/s | 33.0988 Ops/s | $\color{#d91a1a}-1.73\%$ |
| test_iql_speed[True-None] | 8.9080ms | 8.0349ms | 124.4566 Ops/s | 123.5271 Ops/s | $\color{#35bf28}+0.75\%$ |
| test_iql_speed[True-backward] | 17.4441ms | 16.9458ms | 59.0118 Ops/s | 56.8400 Ops/s | $\color{#35bf28}+3.82\%$ |
| test_iql_speed[reduce-overhead-None] | 8.5668ms | 8.1379ms | 122.8814 Ops/s | 125.2144 Ops/s | $\color{#d91a1a}-1.86\%$ |
| test_iql_speed[reduce-overhead-backward] | 18.0542ms | 17.2282ms | 58.0442 Ops/s | 58.1961 Ops/s | $\color{#d91a1a}-0.26\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.2382ms | 6.0528ms | 165.2124 Ops/s | 163.6296 Ops/s | $\color{#35bf28}+0.97\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.9457ms | 0.3281ms | 3.0475 KOps/s | 3.0947 KOps/s | $\color{#d91a1a}-1.52\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.5645ms | 0.3249ms | 3.0782 KOps/s | 3.3967 KOps/s | $\textbf{\color{#d91a1a}-9.38\%}$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.2150ms | 5.9055ms | 169.3323 Ops/s | 167.5738 Ops/s | $\color{#35bf28}+1.05\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7644ms | 0.3359ms | 2.9771 KOps/s | 3.8581 KOps/s | $\textbf{\color{#d91a1a}-22.84\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6586ms | 0.3196ms | 3.1294 KOps/s | 4.6875 KOps/s | $\textbf{\color{#d91a1a}-33.24\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.6682ms | 1.4025ms | 713.0353 Ops/s | 799.4484 Ops/s | $\textbf{\color{#d91a1a}-10.81\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5847ms | 1.3198ms | 757.7144 Ops/s | 858.2820 Ops/s | $\textbf{\color{#d91a1a}-11.72\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.3700ms | 6.1503ms | 162.5927 Ops/s | 162.4639 Ops/s | $\color{#35bf28}+0.08\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.2387ms | 0.4738ms | 2.1106 KOps/s | 2.4304 KOps/s | $\textbf{\color{#d91a1a}-13.16\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7396ms | 0.4318ms | 2.3157 KOps/s | 2.6770 KOps/s | $\textbf{\color{#d91a1a}-13.50\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.3468ms | 6.0964ms | 164.0313 Ops/s | 165.0320 Ops/s | $\color{#d91a1a}-0.61\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.0844ms | 0.2940ms | 3.4008 KOps/s | 2.7922 KOps/s | $\textbf{\color{#35bf28}+21.80\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.4677ms | 0.2422ms | 4.1289 KOps/s | 2.9374 KOps/s | $\textbf{\color{#35bf28}+40.56\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.3577ms | 6.0248ms | 165.9806 Ops/s | 164.3893 Ops/s | $\color{#35bf28}+0.97\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.0353ms | 0.3095ms | 3.2309 KOps/s | 2.8032 KOps/s | $\textbf{\color{#35bf28}+15.26\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5270ms | 0.2170ms | 4.6078 KOps/s | 2.9820 KOps/s | $\textbf{\color{#35bf28}+54.52\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.4245ms | 6.2148ms | 160.9062 Ops/s | 159.7968 Ops/s | $\color{#35bf28}+0.69\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.8262ms | 0.3983ms | 2.5106 KOps/s | 2.2682 KOps/s | $\textbf{\color{#35bf28}+10.69\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.6083ms | 0.3645ms | 2.7434 KOps/s | 2.3416 KOps/s | $\textbf{\color{#35bf28}+17.16\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4473s | 14.2129ms | 70.3587 Ops/s | 187.4898 Ops/s | $\textbf{\color{#d91a1a}-62.47\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 6.8437ms | 2.0794ms | 480.9016 Ops/s | 456.6601 Ops/s | $\textbf{\color{#35bf28}+5.31\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 9.2717ms | 1.3206ms | 757.2238 Ops/s | 830.2348 Ops/s | $\textbf{\color{#d91a1a}-8.79\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 8.4121ms | 5.3600ms | 186.5673 Ops/s | 184.6110 Ops/s | $\color{#35bf28}+1.06\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 8.1925ms | 2.0687ms | 483.4038 Ops/s | 448.9932 Ops/s | $\textbf{\color{#35bf28}+7.66\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 8.2423ms | 1.3033ms | 767.2774 Ops/s | 832.8722 Ops/s | $\textbf{\color{#d91a1a}-7.88\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.4371s | 14.1745ms | 70.5491 Ops/s | 183.4370 Ops/s | $\textbf{\color{#d91a1a}-61.54\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 9.9904ms | 2.2569ms | 443.0807 Ops/s | 402.0714 Ops/s | $\textbf{\color{#35bf28}+10.20\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 7.4774ms | 1.4542ms | 687.6593 Ops/s | 799.0295 Ops/s | $\textbf{\color{#d91a1a}-13.94\%}$ |