rl
rl copied to clipboard
pytorch
[Feature] MCTSForest
:link: Helpful Links
:test_tube: See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2307
- :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, 13 Unrelated Failures
As of commit 570afff294e1eccffb6d0467a4430e4c269a49d5 with merge base 0eabb789739a5e9a2a9f244076c7a9bf8bc7b48e (
NEW FAILURES - The following jobs have failed:
- Habitat Tests on Linux / tests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t af6d1ada3016c1f9830f7d90c9b660a3e5c7d1ddd69823320fdbf324a21ff275 /exec failed with exit code 1 - Unit-tests on Windows / unittests-cpu / windows-job (gh)
##[error]fatal: couldn't find remote ref refs/pull/2307/merge
FLAKY - The following job failed but was likely due to flakiness present on trunk:
- Unit-tests on Linux / tests-cpu-oldget (3.12) / linux-job (gh) (similar failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv]
BROKEN TRUNK - The following jobs failed but was present on the merge base:
👉 Rebase onto the `viable/strict` branch to avoid these failures
- Build Windows Wheels / pytorch/rl / upload / wheel-py3_9-cpu (gh) (trunk failure)
##[error]Unable to find any artifacts for the associated workflow - 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-cpu (3.10) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv] - Unit-tests on Linux / tests-cpu (3.11) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv] - Unit-tests on Linux / tests-cpu (3.12) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv] - Unit-tests on Linux / tests-cpu (3.9) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv] - Unit-tests on Linux / tests-gpu (3.11, 12.1) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv] - Unit-tests on Linux / tests-olddeps (3.8, 11.6) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_spec_shape_inplace_correction - Unit-tests on Linux / tests-optdeps (3.11, 12.1) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_spec_shape_inplace_correction - Unit-tests on Linux / tests-stable-gpu (3.10, 11.8) / linux-job (gh) (trunk failure)
test/test_transforms.py::TestTensorDictPrimer::test_tensordictprimer_batching[False-SerialEnv]
This comment was automatically generated by Dr. CI and updates every 15 minutes.
![pytorch-bot[bot] avatar](https://avatars.githubusercontent.com/in/40112?v=4 "Published by a pub.dev verified publisher"){kind=small}
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests
Total Benchmarks: 149. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}5$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.4286s | 0.4247s | 2.3548 Ops/s | 2.2828 Ops/s | $\color{#35bf28}+3.16\%$ |
| test_transformed | 0.6778s | 0.6069s | 1.6478 Ops/s | 1.7286 Ops/s | $\color{#d91a1a}-4.68\%$ |
| test_serial | 1.3425s | 1.3290s | 0.7525 Ops/s | 0.7453 Ops/s | $\color{#35bf28}+0.96\%$ |
| test_parallel | 1.2853s | 1.2734s | 0.7853 Ops/s | 0.7784 Ops/s | $\color{#35bf28}+0.88\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1386ms | 26.4287μs | 37.8377 KOps/s | 36.5441 KOps/s | $\color{#35bf28}+3.54\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 44.0830μs | 15.3465μs | 65.1613 KOps/s | 62.2122 KOps/s | $\color{#35bf28}+4.74\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 56.5760μs | 14.9999μs | 66.6673 KOps/s | 64.3221 KOps/s | $\color{#35bf28}+3.65\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 33.0920μs | 8.7751μs | 113.9583 KOps/s | 112.2639 KOps/s | $\color{#35bf28}+1.51\%$ |
| test_step_mdp_speed[True-True-False-True-True] | 65.1810μs | 28.4880μs | 35.1025 KOps/s | 34.1705 KOps/s | $\color{#35bf28}+2.73\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 44.7840μs | 16.9865μs | 58.8704 KOps/s | 56.5063 KOps/s | $\color{#35bf28}+4.18\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 62.2560μs | 16.6616μs | 60.0184 KOps/s | 57.9050 KOps/s | $\color{#35bf28}+3.65\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 34.3640μs | 10.4271μs | 95.9041 KOps/s | 93.7539 KOps/s | $\color{#35bf28}+2.29\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 71.9640μs | 30.0351μs | 33.2944 KOps/s | 32.1036 KOps/s | $\color{#35bf28}+3.71\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 61.8860μs | 18.7780μs | 53.2538 KOps/s | 51.1167 KOps/s | $\color{#35bf28}+4.18\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 41.6480μs | 16.8716μs | 59.2711 KOps/s | 58.4490 KOps/s | $\color{#35bf28}+1.41\%$ |
| test_step_mdp_speed[True-False-True-False-False] | 35.4860μs | 10.3637μs | 96.4905 KOps/s | 94.0987 KOps/s | $\color{#35bf28}+2.54\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 72.8250μs | 31.7143μs | 31.5315 KOps/s | 30.8283 KOps/s | $\color{#35bf28}+2.28\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 63.7100μs | 20.1198μs | 49.7022 KOps/s | 47.3859 KOps/s | $\color{#35bf28}+4.89\%$ |
| test_step_mdp_speed[True-False-False-False-True] | 70.0140μs | 18.2015μs | 54.9405 KOps/s | 52.9322 KOps/s | $\color{#35bf28}+3.79\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 33.1320μs | 11.9854μs | 83.4349 KOps/s | 81.0686 KOps/s | $\color{#35bf28}+2.92\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 61.6250μs | 30.1759μs | 33.1391 KOps/s | 32.4256 KOps/s | $\color{#35bf28}+2.20\%$ |
| test_step_mdp_speed[False-True-True-True-False] | 51.8370μs | 18.7790μs | 53.2508 KOps/s | 51.4269 KOps/s | $\color{#35bf28}+3.55\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 45.4450μs | 19.1213μs | 52.2978 KOps/s | 50.5195 KOps/s | $\color{#35bf28}+3.52\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 37.0590μs | 11.6919μs | 85.5295 KOps/s | 83.9397 KOps/s | $\color{#35bf28}+1.89\%$ |
| test_step_mdp_speed[False-True-False-True-True] | 67.7070μs | 31.3355μs | 31.9127 KOps/s | 30.9568 KOps/s | $\color{#35bf28}+3.09\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 58.4090μs | 20.2140μs | 49.4707 KOps/s | 47.3823 KOps/s | $\color{#35bf28}+4.41\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 2.9239ms | 20.7092μs | 48.2878 KOps/s | 47.7150 KOps/s | $\color{#35bf28}+1.20\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 37.3100μs | 13.2114μs | 75.6924 KOps/s | 73.9473 KOps/s | $\color{#35bf28}+2.36\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 75.6010μs | 33.1355μs | 30.1792 KOps/s | 29.2671 KOps/s | $\color{#35bf28}+3.12\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 79.2550μs | 21.8402μs | 45.7871 KOps/s | 44.7816 KOps/s | $\color{#35bf28}+2.25\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 49.9330μs | 20.4802μs | 48.8276 KOps/s | 47.0094 KOps/s | $\color{#35bf28}+3.87\%$ |
| test_step_mdp_speed[False-False-True-False-False] | 43.1810μs | 13.1662μs | 75.9518 KOps/s | 72.9233 KOps/s | $\color{#35bf28}+4.15\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 74.5500μs | 34.0180μs | 29.3962 KOps/s | 28.3241 KOps/s | $\color{#35bf28}+3.79\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 50.8050μs | 22.9641μs | 43.5462 KOps/s | 41.7816 KOps/s | $\color{#35bf28}+4.22\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 60.8640μs | 22.0228μs | 45.4075 KOps/s | 44.6016 KOps/s | $\color{#35bf28}+1.81\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 57.3510μs | 14.4938μs | 68.9949 KOps/s | 66.9301 KOps/s | $\color{#35bf28}+3.09\%$ |
| test_values[generalized_advantage_estimate-True-True] | 9.8273ms | 9.3741ms | 106.6769 Ops/s | 104.4265 Ops/s | $\color{#35bf28}+2.16\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 36.3032ms | 33.4823ms | 29.8665 Ops/s | 30.0004 Ops/s | $\color{#d91a1a}-0.45\%$ |
| test_values[td0_return_estimate-False-False] | 0.2268ms | 0.1721ms | 5.8101 KOps/s | 5.9726 KOps/s | $\color{#d91a1a}-2.72\%$ |
| test_values[td1_return_estimate-False-False] | 23.9256ms | 23.4208ms | 42.6972 Ops/s | 41.1079 Ops/s | $\color{#35bf28}+3.87\%$ |
| test_values[vec_td1_return_estimate-False-False] | 60.3284ms | 34.5642ms | 28.9317 Ops/s | 29.9231 Ops/s | $\color{#d91a1a}-3.31\%$ |
| test_values[td_lambda_return_estimate-True-False] | 35.7640ms | 34.0178ms | 29.3963 Ops/s | 28.8929 Ops/s | $\color{#35bf28}+1.74\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 36.1483ms | 33.5158ms | 29.8367 Ops/s | 29.9258 Ops/s | $\color{#d91a1a}-0.30\%$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 10.5396ms | 8.2065ms | 121.8541 Ops/s | 117.8002 Ops/s | $\color{#35bf28}+3.44\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 3.2493ms | 2.0264ms | 493.4770 Ops/s | 511.4337 Ops/s | $\color{#d91a1a}-3.51\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.5716ms | 0.3584ms | 2.7898 KOps/s | 2.7986 KOps/s | $\color{#d91a1a}-0.31\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 48.5335ms | 45.4473ms | 22.0035 Ops/s | 23.6757 Ops/s | $\textbf{\color{#d91a1a}-7.06\%}$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 4.1332ms | 3.0545ms | 327.3831 Ops/s | 328.1900 Ops/s | $\color{#d91a1a}-0.25\%$ |
| test_dqn_speed[False-None] | 6.3358ms | 1.3357ms | 748.6664 Ops/s | 751.6087 Ops/s | $\color{#d91a1a}-0.39\%$ |
| test_dqn_speed[False-backward] | 1.8751ms | 1.8239ms | 548.2637 Ops/s | 548.8286 Ops/s | $\color{#d91a1a}-0.10\%$ |
| test_dqn_speed[True-None] | 0.6452ms | 0.4598ms | 2.1748 KOps/s | 2.1444 KOps/s | $\color{#35bf28}+1.42\%$ |
| test_dqn_speed[True-backward] | 0.9511ms | 0.8872ms | 1.1272 KOps/s | 985.6774 Ops/s | $\textbf{\color{#35bf28}+14.36\%}$ |
| test_dqn_speed[reduce-overhead-None] | 0.7213ms | 0.4720ms | 2.1187 KOps/s | 2.1391 KOps/s | $\color{#d91a1a}-0.95\%$ |
| test_dqn_speed[reduce-overhead-backward] | 0.9452ms | 0.8882ms | 1.1259 KOps/s | 1.1305 KOps/s | $\color{#d91a1a}-0.41\%$ |
| test_ddpg_speed[False-None] | 3.3797ms | 2.7765ms | 360.1670 Ops/s | 360.1792 Ops/s | $-0.00\%$ |
| test_ddpg_speed[False-backward] | 4.0680ms | 3.9463ms | 253.4016 Ops/s | 256.1994 Ops/s | $\color{#d91a1a}-1.09\%$ |
| test_ddpg_speed[True-None] | 1.2291ms | 1.0118ms | 988.3849 Ops/s | 994.9587 Ops/s | $\color{#d91a1a}-0.66\%$ |
| test_ddpg_speed[True-backward] | 2.3025ms | 2.1682ms | 461.2173 Ops/s | 524.6125 Ops/s | $\textbf{\color{#d91a1a}-12.08\%}$ |
| test_ddpg_speed[reduce-overhead-None] | 1.3498ms | 1.0047ms | 995.3288 Ops/s | 992.7567 Ops/s | $\color{#35bf28}+0.26\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.1394ms | 1.9192ms | 521.0400 Ops/s | 521.1188 Ops/s | $\color{#d91a1a}-0.02\%$ |
| test_sac_speed[False-None] | 8.3827ms | 7.9365ms | 126.0008 Ops/s | 126.4246 Ops/s | $\color{#d91a1a}-0.34\%$ |
| test_sac_speed[False-backward] | 13.8589ms | 10.7677ms | 92.8700 Ops/s | 93.7508 Ops/s | $\color{#d91a1a}-0.94\%$ |
| test_sac_speed[True-None] | 2.4337ms | 1.8312ms | 546.0769 Ops/s | 542.0375 Ops/s | $\color{#35bf28}+0.75\%$ |
| test_sac_speed[True-backward] | 4.5061ms | 3.5785ms | 279.4477 Ops/s | 281.4207 Ops/s | $\color{#d91a1a}-0.70\%$ |
| test_sac_speed[reduce-overhead-None] | 2.1254ms | 1.8309ms | 546.1686 Ops/s | 545.2931 Ops/s | $\color{#35bf28}+0.16\%$ |
| test_sac_speed[reduce-overhead-backward] | 3.6746ms | 3.5279ms | 283.4572 Ops/s | 280.5969 Ops/s | $\color{#35bf28}+1.02\%$ |
| test_redq_speed[False-None] | 15.1179ms | 12.7200ms | 78.6161 Ops/s | 78.7217 Ops/s | $\color{#d91a1a}-0.13\%$ |
| test_redq_speed[False-backward] | 40.9736ms | 22.8860ms | 43.6948 Ops/s | 44.8599 Ops/s | $\color{#d91a1a}-2.60\%$ |
| test_redq_speed[True-None] | 6.1267ms | 4.5970ms | 217.5331 Ops/s | 214.4237 Ops/s | $\color{#35bf28}+1.45\%$ |
| test_redq_speed[True-backward] | 14.0909ms | 12.5807ms | 79.4866 Ops/s | 81.7885 Ops/s | $\color{#d91a1a}-2.81\%$ |
| test_redq_speed[reduce-overhead-None] | 5.4992ms | 4.6293ms | 216.0166 Ops/s | 210.6916 Ops/s | $\color{#35bf28}+2.53\%$ |
| test_redq_speed[reduce-overhead-backward] | 13.5775ms | 12.1538ms | 82.2786 Ops/s | 81.6368 Ops/s | $\color{#35bf28}+0.79\%$ |
| test_redq_deprec_speed[False-None] | 15.5250ms | 12.6845ms | 78.8364 Ops/s | 76.0455 Ops/s | $\color{#35bf28}+3.67\%$ |
| test_redq_deprec_speed[False-backward] | 20.7756ms | 18.6763ms | 53.5439 Ops/s | 54.5239 Ops/s | $\color{#d91a1a}-1.80\%$ |
| test_redq_deprec_speed[True-None] | 4.3992ms | 3.6009ms | 277.7113 Ops/s | 276.3922 Ops/s | $\color{#35bf28}+0.48\%$ |
| test_redq_deprec_speed[True-backward] | 9.0012ms | 8.1149ms | 123.2306 Ops/s | 123.4166 Ops/s | $\color{#d91a1a}-0.15\%$ |
| test_redq_deprec_speed[reduce-overhead-None] | 4.1861ms | 3.5313ms | 283.1823 Ops/s | 275.4686 Ops/s | $\color{#35bf28}+2.80\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 8.5913ms | 7.9581ms | 125.6582 Ops/s | 122.1063 Ops/s | $\color{#35bf28}+2.91\%$ |
| test_td3_speed[False-None] | 8.4546ms | 7.6985ms | 129.8949 Ops/s | 130.0780 Ops/s | $\color{#d91a1a}-0.14\%$ |
| test_td3_speed[False-backward] | 12.7264ms | 10.3007ms | 97.0812 Ops/s | 98.8096 Ops/s | $\color{#d91a1a}-1.75\%$ |
| test_td3_speed[True-None] | 1.9219ms | 1.7095ms | 584.9602 Ops/s | 580.1743 Ops/s | $\color{#35bf28}+0.82\%$ |
| test_td3_speed[True-backward] | 4.0555ms | 3.3226ms | 300.9664 Ops/s | 299.0438 Ops/s | $\color{#35bf28}+0.64\%$ |
| test_td3_speed[reduce-overhead-None] | 1.9185ms | 1.7083ms | 585.3623 Ops/s | 577.9944 Ops/s | $\color{#35bf28}+1.27\%$ |
| test_td3_speed[reduce-overhead-backward] | 3.3486ms | 3.2806ms | 304.8227 Ops/s | 297.5971 Ops/s | $\color{#35bf28}+2.43\%$ |
| test_cql_speed[False-None] | 36.5379ms | 35.0111ms | 28.5624 Ops/s | 27.9565 Ops/s | $\color{#35bf28}+2.17\%$ |
| test_cql_speed[False-backward] | 46.4069ms | 44.8304ms | 22.3063 Ops/s | 21.1666 Ops/s | $\textbf{\color{#35bf28}+5.38\%}$ |
| test_cql_speed[True-None] | 17.1901ms | 15.6477ms | 63.9071 Ops/s | 64.0967 Ops/s | $\color{#d91a1a}-0.30\%$ |
| test_cql_speed[True-backward] | 23.1096ms | 22.1562ms | 45.1341 Ops/s | 44.6779 Ops/s | $\color{#35bf28}+1.02\%$ |
| test_cql_speed[reduce-overhead-None] | 16.8022ms | 15.4251ms | 64.8295 Ops/s | 62.4538 Ops/s | $\color{#35bf28}+3.80\%$ |
| test_cql_speed[reduce-overhead-backward] | 23.7492ms | 22.5456ms | 44.3546 Ops/s | 44.6524 Ops/s | $\color{#d91a1a}-0.67\%$ |
| test_a2c_speed[False-None] | 8.7844ms | 7.0814ms | 141.2145 Ops/s | 140.3469 Ops/s | $\color{#35bf28}+0.62\%$ |
| test_a2c_speed[False-backward] | 15.0778ms | 14.0340ms | 71.2556 Ops/s | 69.7563 Ops/s | $\color{#35bf28}+2.15\%$ |
| test_a2c_speed[True-None] | 3.7320ms | 3.3008ms | 302.9540 Ops/s | 295.3186 Ops/s | $\color{#35bf28}+2.59\%$ |
| test_a2c_speed[True-backward] | 10.6772ms | 9.8031ms | 102.0082 Ops/s | 100.6194 Ops/s | $\color{#35bf28}+1.38\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.9439ms | 3.2987ms | 303.1486 Ops/s | 300.4168 Ops/s | $\color{#35bf28}+0.91\%$ |
| test_a2c_speed[reduce-overhead-backward] | 10.3520ms | 9.8131ms | 101.9048 Ops/s | 100.3011 Ops/s | $\color{#35bf28}+1.60\%$ |
| test_ppo_speed[False-None] | 10.2442ms | 7.4282ms | 134.6225 Ops/s | 132.2807 Ops/s | $\color{#35bf28}+1.77\%$ |
| test_ppo_speed[False-backward] | 17.9058ms | 14.7576ms | 67.7618 Ops/s | 65.4300 Ops/s | $\color{#35bf28}+3.56\%$ |
| test_ppo_speed[True-None] | 3.9457ms | 3.7049ms | 269.9153 Ops/s | 263.1933 Ops/s | $\color{#35bf28}+2.55\%$ |
| test_ppo_speed[True-backward] | 11.9227ms | 9.8287ms | 101.7425 Ops/s | 101.8807 Ops/s | $\color{#d91a1a}-0.14\%$ |
| test_ppo_speed[reduce-overhead-None] | 4.3288ms | 3.6720ms | 272.3305 Ops/s | 265.3307 Ops/s | $\color{#35bf28}+2.64\%$ |
| test_ppo_speed[reduce-overhead-backward] | 10.6434ms | 9.6677ms | 103.4368 Ops/s | 102.1813 Ops/s | $\color{#35bf28}+1.23\%$ |
| test_reinforce_speed[False-None] | 7.2362ms | 6.4312ms | 155.4931 Ops/s | 151.6748 Ops/s | $\color{#35bf28}+2.52\%$ |
| test_reinforce_speed[False-backward] | 10.7167ms | 9.8181ms | 101.8529 Ops/s | 101.2769 Ops/s | $\color{#35bf28}+0.57\%$ |
| test_reinforce_speed[True-None] | 3.2260ms | 2.6244ms | 381.0378 Ops/s | 362.2398 Ops/s | $\textbf{\color{#35bf28}+5.19\%}$ |
| test_reinforce_speed[True-backward] | 8.9788ms | 8.6713ms | 115.3223 Ops/s | 113.7429 Ops/s | $\color{#35bf28}+1.39\%$ |
| test_reinforce_speed[reduce-overhead-None] | 3.6492ms | 2.6387ms | 378.9808 Ops/s | 373.6365 Ops/s | $\color{#35bf28}+1.43\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 8.9508ms | 8.5741ms | 116.6299 Ops/s | 113.4767 Ops/s | $\color{#35bf28}+2.78\%$ |
| test_iql_speed[False-None] | 33.7206ms | 31.6586ms | 31.5869 Ops/s | 30.9758 Ops/s | $\color{#35bf28}+1.97\%$ |
| test_iql_speed[False-backward] | 46.1533ms | 44.7097ms | 22.3665 Ops/s | 21.7929 Ops/s | $\color{#35bf28}+2.63\%$ |
| test_iql_speed[True-None] | 11.4035ms | 10.5285ms | 94.9801 Ops/s | 93.3917 Ops/s | $\color{#35bf28}+1.70\%$ |
| test_iql_speed[True-backward] | 24.3568ms | 21.7967ms | 45.8784 Ops/s | 45.0404 Ops/s | $\color{#35bf28}+1.86\%$ |
| test_iql_speed[reduce-overhead-None] | 11.0009ms | 10.5252ms | 95.0105 Ops/s | 90.9814 Ops/s | $\color{#35bf28}+4.43\%$ |
| test_iql_speed[reduce-overhead-backward] | 22.5146ms | 21.5815ms | 46.3359 Ops/s | 45.6866 Ops/s | $\color{#35bf28}+1.42\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.9519ms | 4.7602ms | 210.0761 Ops/s | 206.1267 Ops/s | $\color{#35bf28}+1.92\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.3689ms | 0.5935ms | 1.6849 KOps/s | 1.5122 KOps/s | $\textbf{\color{#35bf28}+11.42\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.8493ms | 0.6078ms | 1.6454 KOps/s | 1.5942 KOps/s | $\color{#35bf28}+3.21\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.1364ms | 4.5899ms | 217.8678 Ops/s | 214.0254 Ops/s | $\color{#35bf28}+1.80\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.9059ms | 0.6473ms | 1.5449 KOps/s | 1.5715 KOps/s | $\color{#d91a1a}-1.70\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.8287ms | 0.6111ms | 1.6364 KOps/s | 1.6040 KOps/s | $\color{#35bf28}+2.02\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 2.1346ms | 1.8753ms | 533.2344 Ops/s | 529.1930 Ops/s | $\color{#35bf28}+0.76\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 6.3686ms | 1.9170ms | 521.6408 Ops/s | 550.0689 Ops/s | $\textbf{\color{#d91a1a}-5.17\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.1680ms | 4.7098ms | 212.3253 Ops/s | 208.9812 Ops/s | $\color{#35bf28}+1.60\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.5155ms | 0.7813ms | 1.2800 KOps/s | 541.5320 Ops/s | $\textbf{\color{#35bf28}+136.36\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.9469ms | 0.7537ms | 1.3269 KOps/s | 1.3014 KOps/s | $\color{#35bf28}+1.96\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.9522ms | 4.5275ms | 220.8724 Ops/s | 212.8820 Ops/s | $\color{#35bf28}+3.75\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 2.2035ms | 0.5570ms | 1.7952 KOps/s | 1.5093 KOps/s | $\textbf{\color{#35bf28}+18.94\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.8196ms | 0.6086ms | 1.6432 KOps/s | 1.5812 KOps/s | $\color{#35bf28}+3.93\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.0566ms | 4.5595ms | 219.3212 Ops/s | 212.7799 Ops/s | $\color{#35bf28}+3.07\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.1341ms | 0.5985ms | 1.6709 KOps/s | 1.5411 KOps/s | $\textbf{\color{#35bf28}+8.42\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 8.0127ms | 0.6213ms | 1.6095 KOps/s | 1.6234 KOps/s | $\color{#d91a1a}-0.86\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.5864ms | 4.7243ms | 211.6697 Ops/s | 205.9000 Ops/s | $\color{#35bf28}+2.80\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 1.0271ms | 0.7282ms | 1.3733 KOps/s | 1.2547 KOps/s | $\textbf{\color{#35bf28}+9.44\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 8.4026ms | 0.7676ms | 1.3027 KOps/s | 1.2897 KOps/s | $\color{#35bf28}+1.01\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.3885s | 11.8645ms | 84.2852 Ops/s | 235.8372 Ops/s | $\textbf{\color{#d91a1a}-64.26\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 8.2397ms | 2.3324ms | 428.7511 Ops/s | 447.4860 Ops/s | $\color{#d91a1a}-4.19\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 4.7396ms | 1.2527ms | 798.2869 Ops/s | 725.8475 Ops/s | $\textbf{\color{#35bf28}+9.98\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 5.5534ms | 4.1940ms | 238.4375 Ops/s | 236.6211 Ops/s | $\color{#35bf28}+0.77\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 8.0038ms | 2.2930ms | 436.1154 Ops/s | 425.2829 Ops/s | $\color{#35bf28}+2.55\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 4.1242ms | 1.2159ms | 822.4132 Ops/s | 794.0682 Ops/s | $\color{#35bf28}+3.57\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.3673s | 11.5950ms | 86.2441 Ops/s | 237.1212 Ops/s | $\textbf{\color{#d91a1a}-63.63\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 3.5679ms | 2.3043ms | 433.9637 Ops/s | 414.5802 Ops/s | $\color{#35bf28}+4.68\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 1.8271ms | 1.3291ms | 752.3609 Ops/s | 582.9353 Ops/s | $\textbf{\color{#35bf28}+29.06\%}$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] | 11.4997ms | 11.0905ms | 90.1670 Ops/s | 85.9117 Ops/s | $\color{#35bf28}+4.95\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] | 15.6620ms | 14.2774ms | 70.0407 Ops/s | 66.2327 Ops/s | $\textbf{\color{#35bf28}+5.75\%}$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] | 20.7617ms | 19.8733ms | 50.3188 Ops/s | 48.1746 Ops/s | $\color{#35bf28}+4.45\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] | 19.2746ms | 14.3826ms | 69.5285 Ops/s | 67.2015 Ops/s | $\color{#35bf28}+3.46\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] | 20.2307ms | 19.7983ms | 50.5094 Ops/s | 48.6992 Ops/s | $\color{#35bf28}+3.72\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] | 16.3947ms | 15.5159ms | 64.4500 Ops/s | 61.6873 Ops/s | $\color{#35bf28}+4.48\%$ |
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests
Total Benchmarks: 149. Improved: $\large\color{#35bf28}29$. Worsened: $\large\color{#d91a1a}9$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.7760s | 0.7629s | 1.3108 Ops/s | 1.3419 Ops/s | $\color{#d91a1a}-2.32\%$ |
| test_transformed | 1.0857s | 1.0170s | 0.9833 Ops/s | 1.0130 Ops/s | $\color{#d91a1a}-2.93\%$ |
| test_serial | 2.2886s | 2.1592s | 0.4631 Ops/s | 0.4639 Ops/s | $\color{#d91a1a}-0.17\%$ |
| test_parallel | 2.1641s | 2.0930s | 0.4778 Ops/s | 0.5117 Ops/s | $\textbf{\color{#d91a1a}-6.63\%}$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1752ms | 35.1453μs | 28.4533 KOps/s | 26.7520 KOps/s | $\textbf{\color{#35bf28}+6.36\%}$ |
| test_step_mdp_speed[True-True-True-True-False] | 49.9510μs | 20.5832μs | 48.5834 KOps/s | 46.6461 KOps/s | $\color{#35bf28}+4.15\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 49.1410μs | 20.0544μs | 49.8644 KOps/s | 46.5001 KOps/s | $\textbf{\color{#35bf28}+7.23\%}$ |
| test_step_mdp_speed[True-True-True-False-False] | 38.0900μs | 11.6353μs | 85.9456 KOps/s | 81.0606 KOps/s | $\textbf{\color{#35bf28}+6.03\%}$ |
| test_step_mdp_speed[True-True-False-True-True] | 68.4610μs | 37.8803μs | 26.3989 KOps/s | 25.2257 KOps/s | $\color{#35bf28}+4.65\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 47.1000μs | 22.3653μs | 44.7120 KOps/s | 43.0312 KOps/s | $\color{#35bf28}+3.91\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 63.8810μs | 21.7922μs | 45.8880 KOps/s | 41.5805 KOps/s | $\textbf{\color{#35bf28}+10.36\%}$ |
| test_step_mdp_speed[True-True-False-False-False] | 42.7300μs | 13.5879μs | 73.5950 KOps/s | 68.7454 KOps/s | $\textbf{\color{#35bf28}+7.05\%}$ |
| test_step_mdp_speed[True-False-True-True-True] | 76.1420μs | 39.6140μs | 25.2436 KOps/s | 23.8803 KOps/s | $\textbf{\color{#35bf28}+5.71\%}$ |
| test_step_mdp_speed[True-False-True-True-False] | 58.2810μs | 24.1327μs | 41.4375 KOps/s | 38.6289 KOps/s | $\textbf{\color{#35bf28}+7.27\%}$ |
| test_step_mdp_speed[True-False-True-False-True] | 58.9610μs | 21.6235μs | 46.2460 KOps/s | 41.6404 KOps/s | $\textbf{\color{#35bf28}+11.06\%}$ |
| test_step_mdp_speed[True-False-True-False-False] | 55.3410μs | 13.6379μs | 73.3248 KOps/s | 69.9474 KOps/s | $\color{#35bf28}+4.83\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 78.8920μs | 41.7611μs | 23.9457 KOps/s | 22.8551 KOps/s | $\color{#35bf28}+4.77\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 67.5210μs | 26.3478μs | 37.9538 KOps/s | 35.8134 KOps/s | $\textbf{\color{#35bf28}+5.98\%}$ |
| test_step_mdp_speed[True-False-False-False-True] | 51.0610μs | 23.9812μs | 41.6992 KOps/s | 40.0226 KOps/s | $\color{#35bf28}+4.19\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 82.5310μs | 15.6989μs | 63.6986 KOps/s | 61.3407 KOps/s | $\color{#35bf28}+3.84\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 73.1310μs | 40.2420μs | 24.8497 KOps/s | 23.6564 KOps/s | $\textbf{\color{#35bf28}+5.04\%}$ |
| test_step_mdp_speed[False-True-True-True-False] | 47.1110μs | 25.1232μs | 39.8039 KOps/s | 39.0991 KOps/s | $\color{#35bf28}+1.80\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 53.1710μs | 25.9461μs | 38.5415 KOps/s | 36.9305 KOps/s | $\color{#35bf28}+4.36\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 44.1610μs | 15.7581μs | 63.4596 KOps/s | 62.1720 KOps/s | $\color{#35bf28}+2.07\%$ |
| test_step_mdp_speed[False-True-False-True-True] | 73.2910μs | 42.1117μs | 23.7464 KOps/s | 22.8645 KOps/s | $\color{#35bf28}+3.86\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 54.5810μs | 26.1948μs | 38.1756 KOps/s | 36.5450 KOps/s | $\color{#35bf28}+4.46\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 3.3765ms | 28.0205μs | 35.6882 KOps/s | 35.1412 KOps/s | $\color{#35bf28}+1.56\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 43.2610μs | 17.6432μs | 56.6790 KOps/s | 55.4058 KOps/s | $\color{#35bf28}+2.30\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 70.1810μs | 44.5405μs | 22.4515 KOps/s | 21.7249 KOps/s | $\color{#35bf28}+3.34\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 64.6910μs | 28.3784μs | 35.2381 KOps/s | 33.9173 KOps/s | $\color{#35bf28}+3.89\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 62.5810μs | 26.7795μs | 37.3420 KOps/s | 35.4887 KOps/s | $\textbf{\color{#35bf28}+5.22\%}$ |
| test_step_mdp_speed[False-False-True-False-False] | 51.3610μs | 17.2505μs | 57.9694 KOps/s | 56.0297 KOps/s | $\color{#35bf28}+3.46\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 76.7010μs | 44.5385μs | 22.4525 KOps/s | 21.3111 KOps/s | $\textbf{\color{#35bf28}+5.36\%}$ |
| test_step_mdp_speed[False-False-False-True-False] | 62.4510μs | 30.3985μs | 32.8963 KOps/s | 31.3741 KOps/s | $\color{#35bf28}+4.85\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 62.0910μs | 28.9173μs | 34.5814 KOps/s | 33.2287 KOps/s | $\color{#35bf28}+4.07\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 43.3610μs | 19.0461μs | 52.5042 KOps/s | 50.2865 KOps/s | $\color{#35bf28}+4.41\%$ |
| test_values[generalized_advantage_estimate-True-True] | 25.0636ms | 24.5353ms | 40.7575 Ops/s | 41.2806 Ops/s | $\color{#d91a1a}-1.27\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 93.0412ms | 2.7508ms | 363.5297 Ops/s | 330.8364 Ops/s | $\textbf{\color{#35bf28}+9.88\%}$ |
| test_values[td0_return_estimate-False-False] | 85.7820μs | 66.8850μs | 14.9510 KOps/s | 15.0511 KOps/s | $\color{#d91a1a}-0.66\%$ |
| test_values[td1_return_estimate-False-False] | 55.2916ms | 55.0657ms | 18.1601 Ops/s | 18.4319 Ops/s | $\color{#d91a1a}-1.47\%$ |
| test_values[vec_td1_return_estimate-False-False] | 1.3984ms | 1.0780ms | 927.6812 Ops/s | 929.3211 Ops/s | $\color{#d91a1a}-0.18\%$ |
| test_values[td_lambda_return_estimate-True-False] | 88.5991ms | 87.2963ms | 11.4552 Ops/s | 11.5589 Ops/s | $\color{#d91a1a}-0.90\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 1.3960ms | 1.0747ms | 930.4817 Ops/s | 922.5744 Ops/s | $\color{#35bf28}+0.86\%$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.7522ms | 24.2944ms | 41.1617 Ops/s | 41.5614 Ops/s | $\color{#d91a1a}-0.96\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.2167ms | 0.7553ms | 1.3239 KOps/s | 1.3470 KOps/s | $\color{#d91a1a}-1.72\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.7622ms | 0.6611ms | 1.5127 KOps/s | 1.5195 KOps/s | $\color{#d91a1a}-0.45\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.5495ms | 1.4709ms | 679.8367 Ops/s | 678.2938 Ops/s | $\color{#35bf28}+0.23\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.7244ms | 0.6766ms | 1.4779 KOps/s | 1.4890 KOps/s | $\color{#d91a1a}-0.74\%$ |
| test_dqn_speed[False-None] | 6.8030ms | 1.3060ms | 765.7008 Ops/s | 769.1968 Ops/s | $\color{#d91a1a}-0.45\%$ |
| test_dqn_speed[False-backward] | 1.8818ms | 1.8113ms | 552.0891 Ops/s | 544.7726 Ops/s | $\color{#35bf28}+1.34\%$ |
| test_dqn_speed[True-None] | 0.6835ms | 0.5469ms | 1.8283 KOps/s | 1.7313 KOps/s | $\textbf{\color{#35bf28}+5.60\%}$ |
| test_dqn_speed[True-backward] | 1.0400ms | 0.9976ms | 1.0024 KOps/s | 982.8745 Ops/s | $\color{#35bf28}+1.99\%$ |
| test_dqn_speed[reduce-overhead-None] | 0.6678ms | 0.5474ms | 1.8268 KOps/s | 1.7754 KOps/s | $\color{#35bf28}+2.90\%$ |
| test_dqn_speed[reduce-overhead-backward] | 1.0354ms | 1.0031ms | 996.8892 Ops/s | 1.0057 KOps/s | $\color{#d91a1a}-0.87\%$ |
| test_ddpg_speed[False-None] | 3.0675ms | 2.6755ms | 373.7681 Ops/s | 376.3410 Ops/s | $\color{#d91a1a}-0.68\%$ |
| test_ddpg_speed[False-backward] | 3.9564ms | 3.8276ms | 261.2587 Ops/s | 259.1721 Ops/s | $\color{#35bf28}+0.81\%$ |
| test_ddpg_speed[True-None] | 1.3554ms | 1.2259ms | 815.7024 Ops/s | 815.1250 Ops/s | $\color{#35bf28}+0.07\%$ |
| test_ddpg_speed[True-backward] | 2.4588ms | 2.2517ms | 444.1011 Ops/s | 426.4700 Ops/s | $\color{#35bf28}+4.13\%$ |
| test_ddpg_speed[reduce-overhead-None] | 1.3710ms | 1.2273ms | 814.7724 Ops/s | 818.6965 Ops/s | $\color{#d91a1a}-0.48\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.2445ms | 2.1961ms | 455.3561 Ops/s | 456.4054 Ops/s | $\color{#d91a1a}-0.23\%$ |
| test_sac_speed[False-None] | 8.5790ms | 7.4609ms | 134.0315 Ops/s | 133.2527 Ops/s | $\color{#35bf28}+0.58\%$ |
| test_sac_speed[False-backward] | 10.8218ms | 10.5540ms | 94.7505 Ops/s | 94.0089 Ops/s | $\color{#35bf28}+0.79\%$ |
| test_sac_speed[True-None] | 2.3537ms | 1.9823ms | 504.4765 Ops/s | 489.7646 Ops/s | $\color{#35bf28}+3.00\%$ |
| test_sac_speed[True-backward] | 3.9973ms | 3.8649ms | 258.7361 Ops/s | 219.4428 Ops/s | $\textbf{\color{#35bf28}+17.91\%}$ |
| test_sac_speed[reduce-overhead-None] | 2.3756ms | 1.9876ms | 503.1104 Ops/s | 502.3962 Ops/s | $\color{#35bf28}+0.14\%$ |
| test_sac_speed[reduce-overhead-backward] | 4.0864ms | 3.8980ms | 256.5394 Ops/s | 258.3508 Ops/s | $\color{#d91a1a}-0.70\%$ |
| test_redq_speed[False-None] | 16.0655ms | 11.4867ms | 87.0573 Ops/s | 98.7633 Ops/s | $\textbf{\color{#d91a1a}-11.85\%}$ |
| test_redq_speed[False-backward] | 18.0015ms | 17.1979ms | 58.1466 Ops/s | 56.7374 Ops/s | $\color{#35bf28}+2.48\%$ |
| test_redq_speed[True-None] | 3.7916ms | 3.4919ms | 286.3793 Ops/s | 280.2242 Ops/s | $\color{#35bf28}+2.20\%$ |
| test_redq_speed[True-backward] | 8.9436ms | 8.5579ms | 116.8504 Ops/s | 109.9974 Ops/s | $\textbf{\color{#35bf28}+6.23\%}$ |
| test_redq_speed[reduce-overhead-None] | 3.9852ms | 3.5186ms | 284.2032 Ops/s | 295.2866 Ops/s | $\color{#d91a1a}-3.75\%$ |
| test_redq_speed[reduce-overhead-backward] | 8.8244ms | 8.4514ms | 118.3236 Ops/s | 118.5072 Ops/s | $\color{#d91a1a}-0.15\%$ |
| test_redq_deprec_speed[False-None] | 11.3135ms | 10.5417ms | 94.8613 Ops/s | 93.7481 Ops/s | $\color{#35bf28}+1.19\%$ |
| test_redq_deprec_speed[False-backward] | 15.8922ms | 15.2735ms | 65.4728 Ops/s | 65.4837 Ops/s | $\color{#d91a1a}-0.02\%$ |
| test_redq_deprec_speed[True-None] | 3.6992ms | 3.2154ms | 311.0028 Ops/s | 307.3968 Ops/s | $\color{#35bf28}+1.17\%$ |
| test_redq_deprec_speed[True-backward] | 7.4181ms | 7.0946ms | 140.9531 Ops/s | 146.4769 Ops/s | $\color{#d91a1a}-3.77\%$ |
| test_redq_deprec_speed[reduce-overhead-None] | 3.6791ms | 3.2096ms | 311.5693 Ops/s | 317.0675 Ops/s | $\color{#d91a1a}-1.73\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 7.6067ms | 7.1234ms | 140.3815 Ops/s | 139.0505 Ops/s | $\color{#35bf28}+0.96\%$ |
| test_td3_speed[False-None] | 7.5997ms | 7.3755ms | 135.5847 Ops/s | 133.1425 Ops/s | $\color{#35bf28}+1.83\%$ |
| test_td3_speed[False-backward] | 10.6618ms | 10.1801ms | 98.2311 Ops/s | 96.9410 Ops/s | $\color{#35bf28}+1.33\%$ |
| test_td3_speed[True-None] | 1.9336ms | 1.8864ms | 530.1088 Ops/s | 530.3860 Ops/s | $\color{#d91a1a}-0.05\%$ |
| test_td3_speed[True-backward] | 3.8158ms | 3.6448ms | 274.3663 Ops/s | 274.1870 Ops/s | $\color{#35bf28}+0.07\%$ |
| test_td3_speed[reduce-overhead-None] | 1.9899ms | 1.8768ms | 532.8109 Ops/s | 533.7526 Ops/s | $\color{#d91a1a}-0.18\%$ |
| test_td3_speed[reduce-overhead-backward] | 3.7578ms | 3.6752ms | 272.0953 Ops/s | 280.5878 Ops/s | $\color{#d91a1a}-3.03\%$ |
| test_cql_speed[False-None] | 28.6344ms | 24.7349ms | 40.4287 Ops/s | 41.1314 Ops/s | $\color{#d91a1a}-1.71\%$ |
| test_cql_speed[False-backward] | 39.4692ms | 35.0942ms | 28.4948 Ops/s | 30.2152 Ops/s | $\textbf{\color{#d91a1a}-5.69\%}$ |
| test_cql_speed[True-None] | 11.2714ms | 10.8065ms | 92.5370 Ops/s | 93.9855 Ops/s | $\color{#d91a1a}-1.54\%$ |
| test_cql_speed[True-backward] | 16.8766ms | 16.5722ms | 60.3419 Ops/s | 60.8507 Ops/s | $\color{#d91a1a}-0.84\%$ |
| test_cql_speed[reduce-overhead-None] | 11.4802ms | 10.8071ms | 92.5322 Ops/s | 92.5965 Ops/s | $\color{#d91a1a}-0.07\%$ |
| test_cql_speed[reduce-overhead-backward] | 17.3117ms | 16.5338ms | 60.4820 Ops/s | 61.0623 Ops/s | $\color{#d91a1a}-0.95\%$ |
| test_a2c_speed[False-None] | 5.6052ms | 5.2163ms | 191.7068 Ops/s | 185.9726 Ops/s | $\color{#35bf28}+3.08\%$ |
| test_a2c_speed[False-backward] | 11.8900ms | 11.5588ms | 86.5143 Ops/s | 85.5532 Ops/s | $\color{#35bf28}+1.12\%$ |
| test_a2c_speed[True-None] | 3.2423ms | 3.0051ms | 332.7627 Ops/s | 328.1339 Ops/s | $\color{#35bf28}+1.41\%$ |
| test_a2c_speed[True-backward] | 8.8808ms | 8.3532ms | 119.7142 Ops/s | 115.1226 Ops/s | $\color{#35bf28}+3.99\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.1550ms | 2.9882ms | 334.6516 Ops/s | 329.5286 Ops/s | $\color{#35bf28}+1.55\%$ |
| test_a2c_speed[reduce-overhead-backward] | 8.8479ms | 8.3450ms | 119.8328 Ops/s | 120.2982 Ops/s | $\color{#d91a1a}-0.39\%$ |
| test_ppo_speed[False-None] | 6.2222ms | 5.6261ms | 177.7445 Ops/s | 176.5716 Ops/s | $\color{#35bf28}+0.66\%$ |
| test_ppo_speed[False-backward] | 13.7940ms | 12.2066ms | 81.9229 Ops/s | 82.8949 Ops/s | $\color{#d91a1a}-1.17\%$ |
| test_ppo_speed[True-None] | 3.5051ms | 3.3577ms | 297.8261 Ops/s | 292.5105 Ops/s | $\color{#35bf28}+1.82\%$ |
| test_ppo_speed[True-backward] | 8.9476ms | 8.2064ms | 121.8558 Ops/s | 124.0590 Ops/s | $\color{#d91a1a}-1.78\%$ |
| test_ppo_speed[reduce-overhead-None] | 3.7338ms | 3.3748ms | 296.3154 Ops/s | 298.2176 Ops/s | $\color{#d91a1a}-0.64\%$ |
| test_ppo_speed[reduce-overhead-backward] | 8.6610ms | 8.0666ms | 123.9686 Ops/s | 123.2604 Ops/s | $\color{#35bf28}+0.57\%$ |
| test_reinforce_speed[False-None] | 4.9170ms | 4.3884ms | 227.8742 Ops/s | 225.8458 Ops/s | $\color{#35bf28}+0.90\%$ |
| test_reinforce_speed[False-backward] | 7.3577ms | 7.1275ms | 140.3023 Ops/s | 138.4607 Ops/s | $\color{#35bf28}+1.33\%$ |
| test_reinforce_speed[True-None] | 2.3974ms | 2.2071ms | 453.0790 Ops/s | 443.4667 Ops/s | $\color{#35bf28}+2.17\%$ |
| test_reinforce_speed[True-backward] | 7.5458ms | 6.9691ms | 143.4912 Ops/s | 126.6179 Ops/s | $\textbf{\color{#35bf28}+13.33\%}$ |
| test_reinforce_speed[reduce-overhead-None] | 2.6176ms | 2.1893ms | 456.7612 Ops/s | 456.5176 Ops/s | $\color{#35bf28}+0.05\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 7.1497ms | 6.9084ms | 144.7518 Ops/s | 142.7335 Ops/s | $\color{#35bf28}+1.41\%$ |
| test_iql_speed[False-None] | 20.3077ms | 19.3201ms | 51.7595 Ops/s | 50.5605 Ops/s | $\color{#35bf28}+2.37\%$ |
| test_iql_speed[False-backward] | 30.6060ms | 29.7959ms | 33.5617 Ops/s | 32.4728 Ops/s | $\color{#35bf28}+3.35\%$ |
| test_iql_speed[True-None] | 7.3256ms | 6.6822ms | 149.6519 Ops/s | 152.3924 Ops/s | $\color{#d91a1a}-1.80\%$ |
| test_iql_speed[True-backward] | 16.2109ms | 15.2862ms | 65.4186 Ops/s | 65.8585 Ops/s | $\color{#d91a1a}-0.67\%$ |
| test_iql_speed[reduce-overhead-None] | 7.1633ms | 6.6371ms | 150.6685 Ops/s | 151.3657 Ops/s | $\color{#d91a1a}-0.46\%$ |
| test_iql_speed[reduce-overhead-backward] | 16.2458ms | 15.2063ms | 65.7624 Ops/s | 65.7714 Ops/s | $\color{#d91a1a}-0.01\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.4616ms | 6.2875ms | 159.0459 Ops/s | 158.7056 Ops/s | $\color{#35bf28}+0.21\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.8957ms | 0.2702ms | 3.7009 KOps/s | 3.2853 KOps/s | $\textbf{\color{#35bf28}+12.65\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.4901ms | 0.2910ms | 3.4362 KOps/s | 3.1555 KOps/s | $\textbf{\color{#35bf28}+8.90\%}$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.3260ms | 6.0558ms | 165.1318 Ops/s | 165.4333 Ops/s | $\color{#d91a1a}-0.18\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6199ms | 0.2564ms | 3.9003 KOps/s | 3.3073 KOps/s | $\textbf{\color{#35bf28}+17.93\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5632ms | 0.3139ms | 3.1854 KOps/s | 3.6690 KOps/s | $\textbf{\color{#d91a1a}-13.18\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.5386ms | 1.2086ms | 827.4155 Ops/s | 797.6365 Ops/s | $\color{#35bf28}+3.73\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5623ms | 1.1555ms | 865.4263 Ops/s | 820.7665 Ops/s | $\textbf{\color{#35bf28}+5.44\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.2711ms | 6.1662ms | 162.1754 Ops/s | 159.1520 Ops/s | $\color{#35bf28}+1.90\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.6685ms | 0.4027ms | 2.4830 KOps/s | 2.1946 KOps/s | $\textbf{\color{#35bf28}+13.14\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.7161ms | 0.3833ms | 2.6088 KOps/s | 2.3044 KOps/s | $\textbf{\color{#35bf28}+13.21\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.2636ms | 6.1027ms | 163.8624 Ops/s | 164.5650 Ops/s | $\color{#d91a1a}-0.43\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6342ms | 0.2704ms | 3.6984 KOps/s | 3.1318 KOps/s | $\textbf{\color{#35bf28}+18.09\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.4497ms | 0.2486ms | 4.0221 KOps/s | 3.9411 KOps/s | $\color{#35bf28}+2.06\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.2640ms | 6.0397ms | 165.5705 Ops/s | 165.7390 Ops/s | $\color{#d91a1a}-0.10\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.6393ms | 0.2576ms | 3.8813 KOps/s | 3.8364 KOps/s | $\color{#35bf28}+1.17\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6481ms | 0.2373ms | 4.2149 KOps/s | 4.2413 KOps/s | $\color{#d91a1a}-0.62\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.4516ms | 6.2172ms | 160.8445 Ops/s | 161.6701 Ops/s | $\color{#d91a1a}-0.51\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.1261ms | 0.4117ms | 2.4292 KOps/s | 2.1579 KOps/s | $\textbf{\color{#35bf28}+12.57\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.5883ms | 0.3865ms | 2.5875 KOps/s | 2.1287 KOps/s | $\textbf{\color{#35bf28}+21.56\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4144s | 13.3799ms | 74.7387 Ops/s | 186.2290 Ops/s | $\textbf{\color{#d91a1a}-59.87\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 9.6106ms | 1.9767ms | 505.9008 Ops/s | 447.5180 Ops/s | $\textbf{\color{#35bf28}+13.05\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 7.5913ms | 1.2114ms | 825.4671 Ops/s | 957.3813 Ops/s | $\textbf{\color{#d91a1a}-13.78\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 7.1705ms | 5.2176ms | 191.6592 Ops/s | 34.2604 Ops/s | $\textbf{\color{#35bf28}+459.42\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 8.8865ms | 2.0320ms | 492.1280 Ops/s | 619.6989 Ops/s | $\textbf{\color{#d91a1a}-20.59\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 7.1161ms | 1.1662ms | 857.5204 Ops/s | 940.9679 Ops/s | $\textbf{\color{#d91a1a}-8.87\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.3696s | 12.6938ms | 78.7788 Ops/s | 174.8767 Ops/s | $\textbf{\color{#d91a1a}-54.95\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 12.9135ms | 2.0981ms | 476.6295 Ops/s | 457.4617 Ops/s | $\color{#35bf28}+4.19\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 2.3128ms | 1.2139ms | 823.7968 Ops/s | 708.0797 Ops/s | $\textbf{\color{#35bf28}+16.34\%}$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] | 13.6786ms | 12.6990ms | 78.7463 Ops/s | 77.0221 Ops/s | $\color{#35bf28}+2.24\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] | 17.6603ms | 16.4714ms | 60.7113 Ops/s | 60.6618 Ops/s | $\color{#35bf28}+0.08\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] | 18.1550ms | 17.5082ms | 57.1161 Ops/s | 56.6043 Ops/s | $\color{#35bf28}+0.90\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] | 17.2277ms | 16.5632ms | 60.3750 Ops/s | 60.7412 Ops/s | $\color{#d91a1a}-0.60\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] | 17.5359ms | 17.2128ms | 58.0964 Ops/s | 56.9197 Ops/s | $\color{#35bf28}+2.07\%$ |
| test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] | 19.7606ms | 18.1171ms | 55.1964 Ops/s | 56.0565 Ops/s | $\color{#d91a1a}-1.53\%$ |