rl
rl copied to clipboard
[Feature] TensorDictMap hashing functions
:link: Helpful Links
:test_tube: See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2304
- :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, 6 Unrelated Failures
As of commit a08fcde5aedbd0d658540288f082bac27bda2a4e with merge base 194a5ff127eba6fadd945dcf805ad7d004510777 ():
NEW FAILURES - The following jobs have failed:
- Habitat Tests on Linux / tests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t b5ca4eec815522f873c7d25fdcd24a70b65b89677c0c9a9cb83a7c4a85c6c6ae /exec failed with exit code 134 - Lint / c-source / linux-job (gh)
##[error]Process completed with exit code 1.
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-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-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: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}4$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.4142s | 0.4103s | 2.4372 Ops/s | 2.4271 Ops/s | $\color{#35bf28}+0.41\%$ |
| test_transformed | 0.6793s | 0.6037s | 1.6565 Ops/s | 1.7000 Ops/s | $\color{#d91a1a}-2.56\%$ |
| test_serial | 1.4127s | 1.3342s | 0.7495 Ops/s | 0.7442 Ops/s | $\color{#35bf28}+0.71\%$ |
| test_parallel | 1.1936s | 1.1844s | 0.8443 Ops/s | 0.8120 Ops/s | $\color{#35bf28}+3.98\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.2555ms | 28.9284μs | 34.5681 KOps/s | 34.4521 KOps/s | $\color{#35bf28}+0.34\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 70.0530μs | 17.3821μs | 57.5304 KOps/s | 56.9109 KOps/s | $\color{#35bf28}+1.09\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 42.8900μs | 16.0568μs | 62.2787 KOps/s | 61.2437 KOps/s | $\color{#35bf28}+1.69\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 39.5230μs | 9.5888μs | 104.2883 KOps/s | 101.7198 KOps/s | $\color{#35bf28}+2.53\%$ |
| test_step_mdp_speed[True-True-False-True-True] | 86.6120μs | 31.1346μs | 32.1186 KOps/s | 31.9506 KOps/s | $\color{#35bf28}+0.53\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 67.3260μs | 19.4626μs | 51.3806 KOps/s | 50.6597 KOps/s | $\color{#35bf28}+1.42\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 46.7470μs | 18.1022μs | 55.2418 KOps/s | 53.7498 KOps/s | $\color{#35bf28}+2.78\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 52.0870μs | 11.6533μs | 85.8122 KOps/s | 83.3410 KOps/s | $\color{#35bf28}+2.97\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 64.7210μs | 33.4930μs | 29.8570 KOps/s | 29.4658 KOps/s | $\color{#35bf28}+1.33\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 76.4520μs | 21.6310μs | 46.2300 KOps/s | 45.9518 KOps/s | $\color{#35bf28}+0.61\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 52.1180μs | 18.2446μs | 54.8108 KOps/s | 53.5733 KOps/s | $\color{#35bf28}+2.31\%$ |
| test_step_mdp_speed[True-False-True-False-False] | 66.2530μs | 11.6275μs | 86.0030 KOps/s | 84.6111 KOps/s | $\color{#35bf28}+1.65\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 76.0320μs | 35.4589μs | 28.2016 KOps/s | 28.1886 KOps/s | $\color{#35bf28}+0.05\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 61.5350μs | 23.4374μs | 42.6668 KOps/s | 41.9743 KOps/s | $\color{#35bf28}+1.65\%$ |
| test_step_mdp_speed[True-False-False-False-True] | 66.8450μs | 20.1702μs | 49.5781 KOps/s | 49.1456 KOps/s | $\color{#35bf28}+0.88\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 72.1750μs | 13.6186μs | 73.4289 KOps/s | 71.8682 KOps/s | $\color{#35bf28}+2.17\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 91.3510μs | 32.9951μs | 30.3076 KOps/s | 29.6338 KOps/s | $\color{#35bf28}+2.27\%$ |
| test_step_mdp_speed[False-True-True-True-False] | 49.5530μs | 21.5908μs | 46.3159 KOps/s | 45.7862 KOps/s | $\color{#35bf28}+1.16\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 71.2250μs | 20.9842μs | 47.6550 KOps/s | 45.3992 KOps/s | $\color{#35bf28}+4.97\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 2.2699ms | 13.3187μs | 75.0826 KOps/s | 73.3399 KOps/s | $\color{#35bf28}+2.38\%$ |
| test_step_mdp_speed[False-True-False-True-True] | 77.8860μs | 35.1692μs | 28.4340 KOps/s | 27.8800 KOps/s | $\color{#35bf28}+1.99\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 63.1080μs | 23.5260μs | 42.5062 KOps/s | 41.8197 KOps/s | $\color{#35bf28}+1.64\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 86.8830μs | 23.2924μs | 42.9325 KOps/s | 42.5623 KOps/s | $\color{#35bf28}+0.87\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 40.4450μs | 15.3940μs | 64.9605 KOps/s | 64.1533 KOps/s | $\color{#35bf28}+1.26\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 94.0460μs | 37.2578μs | 26.8400 KOps/s | 26.5533 KOps/s | $\color{#35bf28}+1.08\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 58.9400μs | 25.7681μs | 38.8077 KOps/s | 38.9080 KOps/s | $\color{#d91a1a}-0.26\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 78.3160μs | 23.1642μs | 43.1701 KOps/s | 42.0500 KOps/s | $\color{#35bf28}+2.66\%$ |
| test_step_mdp_speed[False-False-True-False-False] | 57.0070μs | 15.2991μs | 65.3633 KOps/s | 63.4360 KOps/s | $\color{#35bf28}+3.04\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 0.2993ms | 39.6213μs | 25.2390 KOps/s | 25.8012 KOps/s | $\color{#d91a1a}-2.18\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 58.8300μs | 27.2520μs | 36.6946 KOps/s | 36.2604 KOps/s | $\color{#35bf28}+1.20\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 77.1500μs | 24.6195μs | 40.6182 KOps/s | 39.4655 KOps/s | $\color{#35bf28}+2.92\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 48.3210μs | 17.1478μs | 58.3167 KOps/s | 57.2450 KOps/s | $\color{#35bf28}+1.87\%$ |
| test_values[generalized_advantage_estimate-True-True] | 9.7820ms | 9.3776ms | 106.6369 Ops/s | 104.1858 Ops/s | $\color{#35bf28}+2.35\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 37.8430ms | 33.6695ms | 29.7005 Ops/s | 27.7027 Ops/s | $\textbf{\color{#35bf28}+7.21\%}$ |
| test_values[td0_return_estimate-False-False] | 0.2274ms | 0.1741ms | 5.7435 KOps/s | 5.7827 KOps/s | $\color{#d91a1a}-0.68\%$ |
| test_values[td1_return_estimate-False-False] | 24.3666ms | 23.6054ms | 42.3633 Ops/s | 41.8929 Ops/s | $\color{#35bf28}+1.12\%$ |
| test_values[vec_td1_return_estimate-False-False] | 34.8636ms | 33.5589ms | 29.7983 Ops/s | 27.6806 Ops/s | $\textbf{\color{#35bf28}+7.65\%}$ |
| test_values[td_lambda_return_estimate-True-False] | 34.6824ms | 33.7576ms | 29.6229 Ops/s | 28.7940 Ops/s | $\color{#35bf28}+2.88\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 35.7676ms | 33.5554ms | 29.8015 Ops/s | 27.6961 Ops/s | $\textbf{\color{#35bf28}+7.60\%}$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 8.4861ms | 8.2229ms | 121.6118 Ops/s | 118.4815 Ops/s | $\color{#35bf28}+2.64\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 2.2918ms | 1.8895ms | 529.2334 Ops/s | 488.1763 Ops/s | $\textbf{\color{#35bf28}+8.41\%}$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.6310ms | 0.3581ms | 2.7922 KOps/s | 2.6925 KOps/s | $\color{#35bf28}+3.70\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 56.0609ms | 48.5156ms | 20.6119 Ops/s | 21.5081 Ops/s | $\color{#d91a1a}-4.17\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 4.7349ms | 3.0784ms | 324.8419 Ops/s | 328.0118 Ops/s | $\color{#d91a1a}-0.97\%$ |
| test_dqn_speed[False-None] | 1.5075ms | 1.3498ms | 740.8636 Ops/s | 724.8638 Ops/s | $\color{#35bf28}+2.21\%$ |
| test_dqn_speed[False-backward] | 1.8860ms | 1.8328ms | 545.6177 Ops/s | 527.9331 Ops/s | $\color{#35bf28}+3.35\%$ |
| test_dqn_speed[True-None] | 1.4111ms | 0.4672ms | 2.1406 KOps/s | 2.1231 KOps/s | $\color{#35bf28}+0.82\%$ |
| test_dqn_speed[True-backward] | 0.9346ms | 0.8835ms | 1.1318 KOps/s | 1.1118 KOps/s | $\color{#35bf28}+1.80\%$ |
| test_dqn_speed[reduce-overhead-None] | 0.7625ms | 0.4723ms | 2.1171 KOps/s | 2.1335 KOps/s | $\color{#d91a1a}-0.76\%$ |
| test_dqn_speed[reduce-overhead-backward] | 0.9411ms | 0.8870ms | 1.1275 KOps/s | 1.1123 KOps/s | $\color{#35bf28}+1.36\%$ |
| test_ddpg_speed[False-None] | 3.6085ms | 2.8029ms | 356.7755 Ops/s | 348.8002 Ops/s | $\color{#35bf28}+2.29\%$ |
| test_ddpg_speed[False-backward] | 4.1769ms | 3.9145ms | 255.4584 Ops/s | 252.2299 Ops/s | $\color{#35bf28}+1.28\%$ |
| test_ddpg_speed[True-None] | 1.3904ms | 1.0117ms | 988.4840 Ops/s | 1.0017 KOps/s | $\color{#d91a1a}-1.32\%$ |
| test_ddpg_speed[True-backward] | 1.9615ms | 1.8995ms | 526.4528 Ops/s | 525.2803 Ops/s | $\color{#35bf28}+0.22\%$ |
| test_ddpg_speed[reduce-overhead-None] | 1.5117ms | 1.0188ms | 981.5235 Ops/s | 1.0071 KOps/s | $\color{#d91a1a}-2.54\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.0059ms | 1.9053ms | 524.8493 Ops/s | 531.6375 Ops/s | $\color{#d91a1a}-1.28\%$ |
| test_sac_speed[False-None] | 12.4232ms | 10.3459ms | 96.6563 Ops/s | 126.0507 Ops/s | $\textbf{\color{#d91a1a}-23.32\%}$ |
| test_sac_speed[False-backward] | 13.6305ms | 10.7068ms | 93.3988 Ops/s | 93.5684 Ops/s | $\color{#d91a1a}-0.18\%$ |
| test_sac_speed[True-None] | 2.4931ms | 1.8638ms | 536.5484 Ops/s | 527.1004 Ops/s | $\color{#35bf28}+1.79\%$ |
| test_sac_speed[True-backward] | 3.9645ms | 3.5732ms | 279.8614 Ops/s | 282.0841 Ops/s | $\color{#d91a1a}-0.79\%$ |
| test_sac_speed[reduce-overhead-None] | 2.3726ms | 1.8746ms | 533.4601 Ops/s | 527.4309 Ops/s | $\color{#35bf28}+1.14\%$ |
| test_sac_speed[reduce-overhead-backward] | 3.7015ms | 3.5547ms | 281.3196 Ops/s | 277.9155 Ops/s | $\color{#35bf28}+1.22\%$ |
| test_redq_speed[False-None] | 14.1031ms | 12.8692ms | 77.7048 Ops/s | 74.0356 Ops/s | $\color{#35bf28}+4.96\%$ |
| test_redq_speed[False-backward] | 24.7092ms | 22.1278ms | 45.1921 Ops/s | 43.5267 Ops/s | $\color{#35bf28}+3.83\%$ |
| test_redq_speed[True-None] | 5.8320ms | 4.9582ms | 201.6843 Ops/s | 203.3777 Ops/s | $\color{#d91a1a}-0.83\%$ |
| test_redq_speed[True-backward] | 12.5325ms | 12.0703ms | 82.8481 Ops/s | 79.5070 Ops/s | $\color{#35bf28}+4.20\%$ |
| test_redq_speed[reduce-overhead-None] | 5.2645ms | 4.5904ms | 217.8452 Ops/s | 193.4693 Ops/s | $\textbf{\color{#35bf28}+12.60\%}$ |
| test_redq_speed[reduce-overhead-backward] | 13.7693ms | 12.0899ms | 82.7138 Ops/s | 80.6657 Ops/s | $\color{#35bf28}+2.54\%$ |
| test_redq_deprec_speed[False-None] | 13.6981ms | 12.6903ms | 78.8003 Ops/s | 76.0161 Ops/s | $\color{#35bf28}+3.66\%$ |
| test_redq_deprec_speed[False-backward] | 22.5392ms | 19.2663ms | 51.9041 Ops/s | 52.3398 Ops/s | $\color{#d91a1a}-0.83\%$ |
| test_redq_deprec_speed[True-None] | 3.9743ms | 3.5866ms | 278.8159 Ops/s | 275.5692 Ops/s | $\color{#35bf28}+1.18\%$ |
| test_redq_deprec_speed[True-backward] | 9.5297ms | 8.0140ms | 124.7817 Ops/s | 114.3923 Ops/s | $\textbf{\color{#35bf28}+9.08\%}$ |
| test_redq_deprec_speed[reduce-overhead-None] | 4.1939ms | 3.5663ms | 280.4020 Ops/s | 269.1637 Ops/s | $\color{#35bf28}+4.18\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 10.5184ms | 8.2443ms | 121.2966 Ops/s | 119.6444 Ops/s | $\color{#35bf28}+1.38\%$ |
| test_td3_speed[False-None] | 8.4485ms | 7.9150ms | 126.3429 Ops/s | 124.0217 Ops/s | $\color{#35bf28}+1.87\%$ |
| test_td3_speed[False-backward] | 12.5944ms | 10.5475ms | 94.8088 Ops/s | 94.8360 Ops/s | $\color{#d91a1a}-0.03\%$ |
| test_td3_speed[True-None] | 2.2008ms | 1.8019ms | 554.9651 Ops/s | 569.3872 Ops/s | $\color{#d91a1a}-2.53\%$ |
| test_td3_speed[True-backward] | 3.7844ms | 3.3991ms | 294.1978 Ops/s | 299.0474 Ops/s | $\color{#d91a1a}-1.62\%$ |
| test_td3_speed[reduce-overhead-None] | 2.1067ms | 1.8259ms | 547.6757 Ops/s | 570.0502 Ops/s | $\color{#d91a1a}-3.92\%$ |
| test_td3_speed[reduce-overhead-backward] | 3.5582ms | 3.3685ms | 296.8721 Ops/s | 294.9478 Ops/s | $\color{#35bf28}+0.65\%$ |
| test_cql_speed[False-None] | 38.4932ms | 35.5498ms | 28.1295 Ops/s | 26.9505 Ops/s | $\color{#35bf28}+4.37\%$ |
| test_cql_speed[False-backward] | 48.9381ms | 45.3728ms | 22.0397 Ops/s | 20.8726 Ops/s | $\textbf{\color{#35bf28}+5.59\%}$ |
| test_cql_speed[True-None] | 17.1617ms | 16.1990ms | 61.7321 Ops/s | 62.3149 Ops/s | $\color{#d91a1a}-0.94\%$ |
| test_cql_speed[True-backward] | 23.7427ms | 22.3126ms | 44.8178 Ops/s | 43.7243 Ops/s | $\color{#35bf28}+2.50\%$ |
| test_cql_speed[reduce-overhead-None] | 17.1493ms | 16.0696ms | 62.2294 Ops/s | 62.9557 Ops/s | $\color{#d91a1a}-1.15\%$ |
| test_cql_speed[reduce-overhead-backward] | 23.8889ms | 22.4603ms | 44.5231 Ops/s | 42.1352 Ops/s | $\textbf{\color{#35bf28}+5.67\%}$ |
| test_a2c_speed[False-None] | 9.3321ms | 7.0926ms | 140.9924 Ops/s | 134.0646 Ops/s | $\textbf{\color{#35bf28}+5.17\%}$ |
| test_a2c_speed[False-backward] | 15.6066ms | 14.2545ms | 70.1533 Ops/s | 69.5907 Ops/s | $\color{#35bf28}+0.81\%$ |
| test_a2c_speed[True-None] | 3.7754ms | 3.3759ms | 296.2214 Ops/s | 294.2916 Ops/s | $\color{#35bf28}+0.66\%$ |
| test_a2c_speed[True-backward] | 11.0850ms | 9.8774ms | 101.2414 Ops/s | 98.3508 Ops/s | $\color{#35bf28}+2.94\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.9340ms | 3.3517ms | 298.3529 Ops/s | 293.6228 Ops/s | $\color{#35bf28}+1.61\%$ |
| test_a2c_speed[reduce-overhead-backward] | 10.9354ms | 10.1045ms | 98.9660 Ops/s | 97.8078 Ops/s | $\color{#35bf28}+1.18\%$ |
| test_ppo_speed[False-None] | 8.8243ms | 7.4990ms | 133.3510 Ops/s | 133.7511 Ops/s | $\color{#d91a1a}-0.30\%$ |
| test_ppo_speed[False-backward] | 15.1061ms | 14.6579ms | 68.2224 Ops/s | 68.4889 Ops/s | $\color{#d91a1a}-0.39\%$ |
| test_ppo_speed[True-None] | 4.1950ms | 3.7535ms | 266.4145 Ops/s | 267.9320 Ops/s | $\color{#d91a1a}-0.57\%$ |
| test_ppo_speed[True-backward] | 11.8399ms | 9.8043ms | 101.9964 Ops/s | 103.0383 Ops/s | $\color{#d91a1a}-1.01\%$ |
| test_ppo_speed[reduce-overhead-None] | 4.4672ms | 3.7698ms | 265.2636 Ops/s | 266.0117 Ops/s | $\color{#d91a1a}-0.28\%$ |
| test_ppo_speed[reduce-overhead-backward] | 12.5892ms | 10.1422ms | 98.5984 Ops/s | 97.5302 Ops/s | $\color{#35bf28}+1.10\%$ |
| test_reinforce_speed[False-None] | 8.2138ms | 6.5279ms | 153.1887 Ops/s | 150.3722 Ops/s | $\color{#35bf28}+1.87\%$ |
| test_reinforce_speed[False-backward] | 11.4761ms | 9.7016ms | 103.0760 Ops/s | 100.0444 Ops/s | $\color{#35bf28}+3.03\%$ |
| test_reinforce_speed[True-None] | 3.2198ms | 2.6571ms | 376.3486 Ops/s | 369.5144 Ops/s | $\color{#35bf28}+1.85\%$ |
| test_reinforce_speed[True-backward] | 17.4246ms | 9.5946ms | 104.2255 Ops/s | 113.0718 Ops/s | $\textbf{\color{#d91a1a}-7.82\%}$ |
| test_reinforce_speed[reduce-overhead-None] | 3.1991ms | 2.6516ms | 377.1304 Ops/s | 370.6279 Ops/s | $\color{#35bf28}+1.75\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 9.4585ms | 8.7867ms | 113.8090 Ops/s | 111.3709 Ops/s | $\color{#35bf28}+2.19\%$ |
| test_iql_speed[False-None] | 33.5525ms | 32.0700ms | 31.1818 Ops/s | 30.1771 Ops/s | $\color{#35bf28}+3.33\%$ |
| test_iql_speed[False-backward] | 48.5333ms | 45.0332ms | 22.2058 Ops/s | 21.6867 Ops/s | $\color{#35bf28}+2.39\%$ |
| test_iql_speed[True-None] | 15.6318ms | 13.9000ms | 71.9424 Ops/s | 72.7933 Ops/s | $\color{#d91a1a}-1.17\%$ |
| test_iql_speed[True-backward] | 25.6142ms | 24.9362ms | 40.1023 Ops/s | 38.6105 Ops/s | $\color{#35bf28}+3.86\%$ |
| test_iql_speed[reduce-overhead-None] | 14.9211ms | 13.9600ms | 71.6331 Ops/s | 73.1579 Ops/s | $\color{#d91a1a}-2.08\%$ |
| test_iql_speed[reduce-overhead-backward] | 26.7524ms | 25.3522ms | 39.4443 Ops/s | 39.4207 Ops/s | $\color{#35bf28}+0.06\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.6958ms | 5.2388ms | 190.8850 Ops/s | 196.3743 Ops/s | $\color{#d91a1a}-2.80\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.8181ms | 0.4962ms | 2.0151 KOps/s | 2.0543 KOps/s | $\color{#d91a1a}-1.91\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7957ms | 0.4680ms | 2.1368 KOps/s | 2.1158 KOps/s | $\color{#35bf28}+0.99\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.3903ms | 5.0546ms | 197.8377 Ops/s | 198.4093 Ops/s | $\color{#d91a1a}-0.29\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.3438ms | 0.4937ms | 2.0255 KOps/s | 2.0400 KOps/s | $\color{#d91a1a}-0.71\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7547ms | 0.4786ms | 2.0894 KOps/s | 2.1916 KOps/s | $\color{#d91a1a}-4.66\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 4.2033ms | 1.5854ms | 630.7453 Ops/s | 618.4632 Ops/s | $\color{#35bf28}+1.99\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.7928ms | 1.5253ms | 655.5925 Ops/s | 622.7889 Ops/s | $\textbf{\color{#35bf28}+5.27\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 8.1268ms | 5.3628ms | 186.4708 Ops/s | 196.1756 Ops/s | $\color{#d91a1a}-4.95\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 3.7783ms | 0.6311ms | 1.5846 KOps/s | 1.5779 KOps/s | $\color{#35bf28}+0.42\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.9208ms | 0.6078ms | 1.6452 KOps/s | 1.6375 KOps/s | $\color{#35bf28}+0.47\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 7.6743ms | 5.0791ms | 196.8871 Ops/s | 201.5184 Ops/s | $\color{#d91a1a}-2.30\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7420ms | 0.4858ms | 2.0583 KOps/s | 2.0303 KOps/s | $\color{#35bf28}+1.38\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 7.8223ms | 0.4867ms | 2.0545 KOps/s | 2.1105 KOps/s | $\color{#d91a1a}-2.65\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.7085ms | 5.2016ms | 192.2472 Ops/s | 204.1901 Ops/s | $\textbf{\color{#d91a1a}-5.85\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 3.0333ms | 0.4982ms | 2.0071 KOps/s | 2.0319 KOps/s | $\color{#d91a1a}-1.22\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.6976ms | 0.4737ms | 2.1109 KOps/s | 2.1881 KOps/s | $\color{#d91a1a}-3.53\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 7.0060ms | 5.2096ms | 191.9540 Ops/s | 191.0070 Ops/s | $\color{#35bf28}+0.50\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 3.4462ms | 0.6393ms | 1.5642 KOps/s | 1.5413 KOps/s | $\color{#35bf28}+1.49\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8580ms | 0.6161ms | 1.6231 KOps/s | 1.6646 KOps/s | $\color{#d91a1a}-2.50\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 5.8320ms | 4.3430ms | 230.2557 Ops/s | 239.9625 Ops/s | $\color{#d91a1a}-4.05\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 5.5823ms | 2.3010ms | 434.6020 Ops/s | 426.7104 Ops/s | $\color{#35bf28}+1.85\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 8.4290ms | 1.3464ms | 742.7301 Ops/s | 789.9074 Ops/s | $\textbf{\color{#d91a1a}-5.97\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.4797s | 13.9299ms | 71.7882 Ops/s | 33.7984 Ops/s | $\textbf{\color{#35bf28}+112.40\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 7.0203ms | 2.3311ms | 428.9814 Ops/s | 438.8800 Ops/s | $\color{#d91a1a}-2.26\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 6.0206ms | 1.3046ms | 766.4996 Ops/s | 755.4423 Ops/s | $\color{#35bf28}+1.46\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 5.8897ms | 4.4968ms | 222.3794 Ops/s | 217.3563 Ops/s | $\color{#35bf28}+2.31\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 6.9547ms | 2.5066ms | 398.9485 Ops/s | 393.6552 Ops/s | $\color{#35bf28}+1.34\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 5.2698ms | 1.4791ms | 676.0704 Ops/s | 645.0821 Ops/s | $\color{#35bf28}+4.80\%$ |
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests
Total Benchmarks: 143. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}8$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_simple | 0.7413s | 0.7408s | 1.3498 Ops/s | 1.3428 Ops/s | $\color{#35bf28}+0.52\%$ |
| test_transformed | 1.0800s | 1.0036s | 0.9964 Ops/s | 1.0081 Ops/s | $\color{#d91a1a}-1.16\%$ |
| test_serial | 2.2451s | 2.1683s | 0.4612 Ops/s | 0.4628 Ops/s | $\color{#d91a1a}-0.35\%$ |
| test_parallel | 1.9952s | 1.9090s | 0.5238 Ops/s | 0.5074 Ops/s | $\color{#35bf28}+3.23\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1429ms | 39.1768μs | 25.5253 KOps/s | 25.0904 KOps/s | $\color{#35bf28}+1.73\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 0.1721ms | 22.8615μs | 43.7417 KOps/s | 43.0267 KOps/s | $\color{#35bf28}+1.66\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 0.1464ms | 20.7334μs | 48.2314 KOps/s | 47.0299 KOps/s | $\color{#35bf28}+2.55\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 75.1910μs | 11.7620μs | 85.0192 KOps/s | 79.7976 KOps/s | $\textbf{\color{#35bf28}+6.54\%}$ |
| test_step_mdp_speed[True-True-False-True-True] | 96.8010μs | 41.8352μs | 23.9033 KOps/s | 23.0662 KOps/s | $\color{#35bf28}+3.63\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 66.6010μs | 25.1714μs | 39.7276 KOps/s | 39.0192 KOps/s | $\color{#35bf28}+1.82\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 47.8510μs | 23.8058μs | 42.0066 KOps/s | 40.5898 KOps/s | $\color{#35bf28}+3.49\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 41.6700μs | 15.0116μs | 66.6152 KOps/s | 65.5656 KOps/s | $\color{#35bf28}+1.60\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 86.0310μs | 44.5555μs | 22.4439 KOps/s | 22.4970 KOps/s | $\color{#d91a1a}-0.24\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 62.8010μs | 28.3981μs | 35.2137 KOps/s | 35.1920 KOps/s | $\color{#35bf28}+0.06\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 60.9210μs | 24.1972μs | 41.3271 KOps/s | 40.1484 KOps/s | $\color{#35bf28}+2.94\%$ |
| test_step_mdp_speed[True-False-True-False-False] | 98.8010μs | 15.2517μs | 65.5667 KOps/s | 65.2006 KOps/s | $\color{#35bf28}+0.56\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 0.2205ms | 47.1707μs | 21.1996 KOps/s | 20.8555 KOps/s | $\color{#35bf28}+1.65\%$ |
| test_step_mdp_speed[True-False-False-True-False] | 0.1168ms | 31.2672μs | 31.9824 KOps/s | 32.5681 KOps/s | $\color{#d91a1a}-1.80\%$ |
| test_step_mdp_speed[True-False-False-False-True] | 95.3710μs | 26.7152μs | 37.4319 KOps/s | 36.8826 KOps/s | $\color{#35bf28}+1.49\%$ |
| test_step_mdp_speed[True-False-False-False-False] | 62.4200μs | 17.8317μs | 56.0800 KOps/s | 56.4077 KOps/s | $\color{#d91a1a}-0.58\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 77.6310μs | 43.7789μs | 22.8420 KOps/s | 21.9262 KOps/s | $\color{#35bf28}+4.18\%$ |
| test_step_mdp_speed[False-True-True-True-False] | 66.3810μs | 28.4168μs | 35.1905 KOps/s | 34.5561 KOps/s | $\color{#35bf28}+1.84\%$ |
| test_step_mdp_speed[False-True-True-False-True] | 0.1329ms | 28.0484μs | 35.6527 KOps/s | 34.0834 KOps/s | $\color{#35bf28}+4.60\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 2.7059ms | 17.6446μs | 56.6746 KOps/s | 49.5590 KOps/s | $\textbf{\color{#35bf28}+14.36\%}$ |
| test_step_mdp_speed[False-True-False-True-True] | 92.2910μs | 47.4220μs | 21.0873 KOps/s | 20.4469 KOps/s | $\color{#35bf28}+3.13\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 61.4610μs | 30.9781μs | 32.2809 KOps/s | 32.0873 KOps/s | $\color{#35bf28}+0.60\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 0.1454ms | 30.9245μs | 32.3368 KOps/s | 31.0988 KOps/s | $\color{#35bf28}+3.98\%$ |
| test_step_mdp_speed[False-True-False-False-False] | 0.1588ms | 20.0976μs | 49.7571 KOps/s | 49.1925 KOps/s | $\color{#35bf28}+1.15\%$ |
| test_step_mdp_speed[False-False-True-True-True] | 0.2379ms | 50.5208μs | 19.7938 KOps/s | 19.7014 KOps/s | $\color{#35bf28}+0.47\%$ |
| test_step_mdp_speed[False-False-True-True-False] | 59.7410μs | 33.5732μs | 29.7856 KOps/s | 29.5423 KOps/s | $\color{#35bf28}+0.82\%$ |
| test_step_mdp_speed[False-False-True-False-True] | 66.0510μs | 31.7744μs | 31.4719 KOps/s | 31.9183 KOps/s | $\color{#d91a1a}-1.40\%$ |
| test_step_mdp_speed[False-False-True-False-False] | 0.1966ms | 20.4675μs | 48.8578 KOps/s | 49.2183 KOps/s | $\color{#d91a1a}-0.73\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 99.3810μs | 52.6948μs | 18.9772 KOps/s | 18.7901 KOps/s | $\color{#35bf28}+1.00\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 62.3910μs | 36.8167μs | 27.1616 KOps/s | 27.8024 KOps/s | $\color{#d91a1a}-2.30\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 60.1710μs | 33.0417μs | 30.2648 KOps/s | 29.1090 KOps/s | $\color{#35bf28}+3.97\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 0.2104ms | 22.7536μs | 43.9491 KOps/s | 43.8963 KOps/s | $\color{#35bf28}+0.12\%$ |
| test_values[generalized_advantage_estimate-True-True] | 25.8242ms | 25.1107ms | 39.8237 Ops/s | 39.8410 Ops/s | $\color{#d91a1a}-0.04\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 0.1059s | 3.0187ms | 331.2641 Ops/s | 300.8134 Ops/s | $\textbf{\color{#35bf28}+10.12\%}$ |
| test_values[td0_return_estimate-False-False] | 87.8810μs | 67.1144μs | 14.8999 KOps/s | 14.7199 KOps/s | $\color{#35bf28}+1.22\%$ |
| test_values[td1_return_estimate-False-False] | 56.7535ms | 56.0994ms | 17.8255 Ops/s | 17.9179 Ops/s | $\color{#d91a1a}-0.52\%$ |
| test_values[vec_td1_return_estimate-False-False] | 1.3655ms | 1.0887ms | 918.5036 Ops/s | 920.2736 Ops/s | $\color{#d91a1a}-0.19\%$ |
| test_values[td_lambda_return_estimate-True-False] | 90.3738ms | 88.8899ms | 11.2499 Ops/s | 11.2805 Ops/s | $\color{#d91a1a}-0.27\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 1.3436ms | 1.0847ms | 921.9302 Ops/s | 924.8526 Ops/s | $\color{#d91a1a}-0.32\%$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 25.3282ms | 25.0063ms | 39.9900 Ops/s | 39.9047 Ops/s | $\color{#35bf28}+0.21\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 1.0730ms | 0.7524ms | 1.3290 KOps/s | 1.3337 KOps/s | $\color{#d91a1a}-0.35\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.8417ms | 0.6717ms | 1.4888 KOps/s | 1.4946 KOps/s | $\color{#d91a1a}-0.39\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6262ms | 1.4804ms | 675.4727 Ops/s | 677.1360 Ops/s | $\color{#d91a1a}-0.25\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.8650ms | 0.6866ms | 1.4565 KOps/s | 1.4608 KOps/s | $\color{#d91a1a}-0.29\%$ |
| test_dqn_speed[False-None] | 6.5418ms | 1.3394ms | 746.5817 Ops/s | 736.2201 Ops/s | $\color{#35bf28}+1.41\%$ |
| test_dqn_speed[False-backward] | 1.9664ms | 1.8383ms | 543.9841 Ops/s | 528.0857 Ops/s | $\color{#35bf28}+3.01\%$ |
| test_dqn_speed[True-None] | 1.1719ms | 0.5719ms | 1.7487 KOps/s | 1.7169 KOps/s | $\color{#35bf28}+1.85\%$ |
| test_dqn_speed[True-backward] | 1.0760ms | 1.0042ms | 995.8284 Ops/s | 802.9412 Ops/s | $\textbf{\color{#35bf28}+24.02\%}$ |
| test_dqn_speed[reduce-overhead-None] | 0.8645ms | 0.5554ms | 1.8004 KOps/s | 1.6882 KOps/s | $\textbf{\color{#35bf28}+6.65\%}$ |
| test_dqn_speed[reduce-overhead-backward] | 1.0648ms | 1.0114ms | 988.6876 Ops/s | 969.8556 Ops/s | $\color{#35bf28}+1.94\%$ |
| test_ddpg_speed[False-None] | 3.3262ms | 2.7366ms | 365.4222 Ops/s | 360.1847 Ops/s | $\color{#35bf28}+1.45\%$ |
| test_ddpg_speed[False-backward] | 4.2437ms | 3.9599ms | 252.5333 Ops/s | 248.4247 Ops/s | $\color{#35bf28}+1.65\%$ |
| test_ddpg_speed[True-None] | 1.4116ms | 1.2307ms | 812.5618 Ops/s | 787.7005 Ops/s | $\color{#35bf28}+3.16\%$ |
| test_ddpg_speed[True-backward] | 2.3593ms | 2.2455ms | 445.3326 Ops/s | 442.4246 Ops/s | $\color{#35bf28}+0.66\%$ |
| test_ddpg_speed[reduce-overhead-None] | 1.4824ms | 1.2828ms | 779.5579 Ops/s | 785.1190 Ops/s | $\color{#d91a1a}-0.71\%$ |
| test_ddpg_speed[reduce-overhead-backward] | 2.4166ms | 2.2175ms | 450.9510 Ops/s | 443.7179 Ops/s | $\color{#35bf28}+1.63\%$ |
| test_sac_speed[False-None] | 7.9708ms | 7.6406ms | 130.8789 Ops/s | 127.0662 Ops/s | $\color{#35bf28}+3.00\%$ |
| test_sac_speed[False-backward] | 11.2476ms | 10.8634ms | 92.0519 Ops/s | 90.0877 Ops/s | $\color{#35bf28}+2.18\%$ |
| test_sac_speed[True-None] | 2.2352ms | 2.0351ms | 491.3673 Ops/s | 478.7251 Ops/s | $\color{#35bf28}+2.64\%$ |
| test_sac_speed[True-backward] | 4.1927ms | 3.9732ms | 251.6849 Ops/s | 247.7419 Ops/s | $\color{#35bf28}+1.59\%$ |
| test_sac_speed[reduce-overhead-None] | 2.3052ms | 2.0373ms | 490.8489 Ops/s | 481.4213 Ops/s | $\color{#35bf28}+1.96\%$ |
| test_sac_speed[reduce-overhead-backward] | 4.1655ms | 3.9798ms | 251.2680 Ops/s | 250.5096 Ops/s | $\color{#35bf28}+0.30\%$ |
| test_redq_speed[False-None] | 10.7894ms | 9.9486ms | 100.5167 Ops/s | 95.8801 Ops/s | $\color{#35bf28}+4.84\%$ |
| test_redq_speed[False-backward] | 18.0289ms | 17.0466ms | 58.6627 Ops/s | 56.3231 Ops/s | $\color{#35bf28}+4.15\%$ |
| test_redq_speed[True-None] | 3.9134ms | 3.6225ms | 276.0508 Ops/s | 281.7651 Ops/s | $\color{#d91a1a}-2.03\%$ |
| test_redq_speed[True-backward] | 9.0550ms | 8.6091ms | 116.1567 Ops/s | 117.2510 Ops/s | $\color{#d91a1a}-0.93\%$ |
| test_redq_speed[reduce-overhead-None] | 4.0079ms | 3.5528ms | 281.4701 Ops/s | 283.9984 Ops/s | $\color{#d91a1a}-0.89\%$ |
| test_redq_speed[reduce-overhead-backward] | 9.0063ms | 8.5261ms | 117.2869 Ops/s | 118.7234 Ops/s | $\color{#d91a1a}-1.21\%$ |
| test_redq_deprec_speed[False-None] | 12.3062ms | 10.5552ms | 94.7403 Ops/s | 91.4341 Ops/s | $\color{#35bf28}+3.62\%$ |
| test_redq_deprec_speed[False-backward] | 16.0923ms | 15.3672ms | 65.0735 Ops/s | 63.4168 Ops/s | $\color{#35bf28}+2.61\%$ |
| test_redq_deprec_speed[True-None] | 3.5042ms | 3.2385ms | 308.7836 Ops/s | 300.6644 Ops/s | $\color{#35bf28}+2.70\%$ |
| test_redq_deprec_speed[True-backward] | 7.5124ms | 7.1830ms | 139.2175 Ops/s | 144.0065 Ops/s | $\color{#d91a1a}-3.33\%$ |
| test_redq_deprec_speed[reduce-overhead-None] | 3.4698ms | 3.2449ms | 308.1792 Ops/s | 310.0604 Ops/s | $\color{#d91a1a}-0.61\%$ |
| test_redq_deprec_speed[reduce-overhead-backward] | 7.3763ms | 7.1731ms | 139.4089 Ops/s | 145.1628 Ops/s | $\color{#d91a1a}-3.96\%$ |
| test_td3_speed[False-None] | 7.9015ms | 7.6015ms | 131.5535 Ops/s | 126.6163 Ops/s | $\color{#35bf28}+3.90\%$ |
| test_td3_speed[False-backward] | 10.9364ms | 10.4309ms | 95.8689 Ops/s | 92.2601 Ops/s | $\color{#35bf28}+3.91\%$ |
| test_td3_speed[True-None] | 1.9632ms | 1.9215ms | 520.4305 Ops/s | 515.8051 Ops/s | $\color{#35bf28}+0.90\%$ |
| test_td3_speed[True-backward] | 3.9133ms | 3.7472ms | 266.8672 Ops/s | 218.7899 Ops/s | $\textbf{\color{#35bf28}+21.97\%}$ |
| test_td3_speed[reduce-overhead-None] | 1.9682ms | 1.9176ms | 521.4785 Ops/s | 518.4036 Ops/s | $\color{#35bf28}+0.59\%$ |
| test_td3_speed[reduce-overhead-backward] | 3.9411ms | 3.7685ms | 265.3555 Ops/s | 265.6279 Ops/s | $\color{#d91a1a}-0.10\%$ |
| test_cql_speed[False-None] | 29.0061ms | 25.5137ms | 39.1946 Ops/s | 39.1551 Ops/s | $\color{#35bf28}+0.10\%$ |
| test_cql_speed[False-backward] | 38.6887ms | 34.9799ms | 28.5878 Ops/s | 28.4289 Ops/s | $\color{#35bf28}+0.56\%$ |
| test_cql_speed[True-None] | 11.2550ms | 10.9309ms | 91.4841 Ops/s | 91.2735 Ops/s | $\color{#35bf28}+0.23\%$ |
| test_cql_speed[True-backward] | 17.7935ms | 16.9851ms | 58.8750 Ops/s | 61.1684 Ops/s | $\color{#d91a1a}-3.75\%$ |
| test_cql_speed[reduce-overhead-None] | 11.3647ms | 10.9786ms | 91.0863 Ops/s | 92.1069 Ops/s | $\color{#d91a1a}-1.11\%$ |
| test_cql_speed[reduce-overhead-backward] | 17.2074ms | 16.7867ms | 59.5709 Ops/s | 60.5983 Ops/s | $\color{#d91a1a}-1.70\%$ |
| test_a2c_speed[False-None] | 7.3883ms | 5.2323ms | 191.1188 Ops/s | 184.3989 Ops/s | $\color{#35bf28}+3.64\%$ |
| test_a2c_speed[False-backward] | 12.0315ms | 11.6682ms | 85.7031 Ops/s | 85.3247 Ops/s | $\color{#35bf28}+0.44\%$ |
| test_a2c_speed[True-None] | 3.4202ms | 3.0833ms | 324.3321 Ops/s | 318.9986 Ops/s | $\color{#35bf28}+1.67\%$ |
| test_a2c_speed[True-backward] | 8.7399ms | 8.5252ms | 117.2995 Ops/s | 112.4135 Ops/s | $\color{#35bf28}+4.35\%$ |
| test_a2c_speed[reduce-overhead-None] | 3.2285ms | 3.0421ms | 328.7161 Ops/s | 327.4196 Ops/s | $\color{#35bf28}+0.40\%$ |
| test_a2c_speed[reduce-overhead-backward] | 8.7122ms | 8.4286ms | 118.6437 Ops/s | 114.9352 Ops/s | $\color{#35bf28}+3.23\%$ |
| test_ppo_speed[False-None] | 5.9353ms | 5.6501ms | 176.9867 Ops/s | 170.0664 Ops/s | $\color{#35bf28}+4.07\%$ |
| test_ppo_speed[False-backward] | 12.7942ms | 12.2391ms | 81.7051 Ops/s | 79.1436 Ops/s | $\color{#35bf28}+3.24\%$ |
| test_ppo_speed[True-None] | 3.6579ms | 3.4440ms | 290.3602 Ops/s | 289.8881 Ops/s | $\color{#35bf28}+0.16\%$ |
| test_ppo_speed[True-backward] | 8.6773ms | 8.3126ms | 120.2993 Ops/s | 115.9224 Ops/s | $\color{#35bf28}+3.78\%$ |
| test_ppo_speed[reduce-overhead-None] | 3.6235ms | 3.4575ms | 289.2299 Ops/s | 289.1049 Ops/s | $\color{#35bf28}+0.04\%$ |
| test_ppo_speed[reduce-overhead-backward] | 8.4231ms | 8.1753ms | 122.3196 Ops/s | 120.6822 Ops/s | $\color{#35bf28}+1.36\%$ |
| test_reinforce_speed[False-None] | 6.5155ms | 4.4715ms | 223.6378 Ops/s | 218.7334 Ops/s | $\color{#35bf28}+2.24\%$ |
| test_reinforce_speed[False-backward] | 7.4740ms | 7.2842ms | 137.2830 Ops/s | 135.9753 Ops/s | $\color{#35bf28}+0.96\%$ |
| test_reinforce_speed[True-None] | 2.4775ms | 2.2456ms | 445.3172 Ops/s | 441.2867 Ops/s | $\color{#35bf28}+0.91\%$ |
| test_reinforce_speed[True-backward] | 7.3707ms | 7.1245ms | 140.3612 Ops/s | 138.9362 Ops/s | $\color{#35bf28}+1.03\%$ |
| test_reinforce_speed[reduce-overhead-None] | 2.4826ms | 2.2371ms | 446.9984 Ops/s | 448.0977 Ops/s | $\color{#d91a1a}-0.25\%$ |
| test_reinforce_speed[reduce-overhead-backward] | 7.2820ms | 7.0471ms | 141.9030 Ops/s | 140.4239 Ops/s | $\color{#35bf28}+1.05\%$ |
| test_iql_speed[False-None] | 24.6068ms | 20.1002ms | 49.7507 Ops/s | 51.3130 Ops/s | $\color{#d91a1a}-3.04\%$ |
| test_iql_speed[False-backward] | 31.2864ms | 30.0239ms | 33.3068 Ops/s | 33.8434 Ops/s | $\color{#d91a1a}-1.59\%$ |
| test_iql_speed[True-None] | 8.2573ms | 7.8945ms | 126.6707 Ops/s | 126.5261 Ops/s | $\color{#35bf28}+0.11\%$ |
| test_iql_speed[True-backward] | 17.5816ms | 16.6693ms | 59.9906 Ops/s | 59.5850 Ops/s | $\color{#35bf28}+0.68\%$ |
| test_iql_speed[reduce-overhead-None] | 8.3847ms | 7.9176ms | 126.3009 Ops/s | 124.4983 Ops/s | $\color{#35bf28}+1.45\%$ |
| test_iql_speed[reduce-overhead-backward] | 17.2105ms | 16.5727ms | 60.3404 Ops/s | 59.4649 Ops/s | $\color{#35bf28}+1.47\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 6.5339ms | 6.3541ms | 157.3783 Ops/s | 159.7815 Ops/s | $\color{#d91a1a}-1.50\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.6499ms | 0.2408ms | 4.1531 KOps/s | 4.0266 KOps/s | $\color{#35bf28}+3.14\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.5747ms | 0.3152ms | 3.1726 KOps/s | 4.4085 KOps/s | $\textbf{\color{#d91a1a}-28.03\%}$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.6227ms | 6.2671ms | 159.5635 Ops/s | 160.2419 Ops/s | $\color{#d91a1a}-0.42\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7340ms | 0.3125ms | 3.1996 KOps/s | 2.9786 KOps/s | $\textbf{\color{#35bf28}+7.42\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.5415ms | 0.2947ms | 3.3938 KOps/s | 3.5816 KOps/s | $\textbf{\color{#d91a1a}-5.24\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 1.6200ms | 1.3066ms | 765.3350 Ops/s | 747.6470 Ops/s | $\color{#35bf28}+2.37\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 1.5726ms | 1.2523ms | 798.5432 Ops/s | 789.6080 Ops/s | $\color{#35bf28}+1.13\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.6475ms | 6.4520ms | 154.9914 Ops/s | 152.1710 Ops/s | $\color{#35bf28}+1.85\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.3820s | 0.6851ms | 1.4596 KOps/s | 2.1936 KOps/s | $\textbf{\color{#d91a1a}-33.46\%}$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.6351ms | 0.3688ms | 2.7117 KOps/s | 2.3188 KOps/s | $\textbf{\color{#35bf28}+16.94\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 9.8955ms | 6.5198ms | 153.3795 Ops/s | 155.4701 Ops/s | $\color{#d91a1a}-1.34\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.0670ms | 0.3412ms | 2.9306 KOps/s | 3.4210 KOps/s | $\textbf{\color{#d91a1a}-14.34\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.6045ms | 0.2146ms | 4.6601 KOps/s | 3.6792 KOps/s | $\textbf{\color{#35bf28}+26.66\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 6.7004ms | 6.2671ms | 159.5631 Ops/s | 156.7591 Ops/s | $\color{#35bf28}+1.79\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.7268ms | 0.3333ms | 2.9999 KOps/s | 3.5679 KOps/s | $\textbf{\color{#d91a1a}-15.92\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.4326ms | 0.2143ms | 4.6665 KOps/s | 4.0544 KOps/s | $\textbf{\color{#35bf28}+15.10\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 6.6239ms | 6.4655ms | 154.6665 Ops/s | 152.0731 Ops/s | $\color{#35bf28}+1.71\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 2.2588ms | 0.4745ms | 2.1076 KOps/s | 2.3493 KOps/s | $\textbf{\color{#d91a1a}-10.29\%}$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.5567ms | 0.3590ms | 2.7854 KOps/s | 2.4251 KOps/s | $\textbf{\color{#35bf28}+14.86\%}$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.4193s | 13.6991ms | 72.9977 Ops/s | 180.4857 Ops/s | $\textbf{\color{#d91a1a}-59.55\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 7.7930ms | 2.0376ms | 490.7718 Ops/s | 440.4308 Ops/s | $\textbf{\color{#35bf28}+11.43\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 8.2277ms | 1.1991ms | 833.9890 Ops/s | 816.9847 Ops/s | $\color{#35bf28}+2.08\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 7.5256ms | 5.4160ms | 184.6389 Ops/s | 183.0647 Ops/s | $\color{#35bf28}+0.86\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 8.9763ms | 2.0087ms | 497.8241 Ops/s | 420.7737 Ops/s | $\textbf{\color{#35bf28}+18.31\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 7.5304ms | 1.1977ms | 834.9494 Ops/s | 801.8892 Ops/s | $\color{#35bf28}+4.12\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.3789s | 13.0807ms | 76.4484 Ops/s | 178.7659 Ops/s | $\textbf{\color{#d91a1a}-57.24\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 11.0390ms | 2.2857ms | 437.4944 Ops/s | 396.1202 Ops/s | $\textbf{\color{#35bf28}+10.44\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 7.1208ms | 1.3846ms | 722.2492 Ops/s | 736.1434 Ops/s | $\color{#d91a1a}-1.89\%$ |