Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Doc] Add Tokenizer and auto-reset doc link #2754

Merged
merged 1 commit into from
Feb 4, 2025

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 4, 2025

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 4, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2754

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

This comment was automatically generated by Dr. CI and updates every 15 minutes.

vmoens added a commit that referenced this pull request Feb 4, 2025
ghstack-source-id: 90f55b568e85ae151bea4370025144c19e74602b
Pull Request resolved: #2754
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 4, 2025
@vmoens vmoens merged commit 2f0cc4f into gh/vmoens/85/base Feb 4, 2025
53 of 61 checks passed
vmoens added a commit that referenced this pull request Feb 4, 2025
ghstack-source-id: 90f55b568e85ae151bea4370025144c19e74602b
Pull Request resolved: #2754
@vmoens vmoens deleted the gh/vmoens/85/head branch February 4, 2025 08:53
Copy link

github-actions bot commented Feb 4, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5531s 0.4591s 2.1783 Ops/s 2.1743 Ops/s $\color{#35bf28}+0.18\%$
test_transformed 1.0560s 0.9559s 1.0462 Ops/s 1.0767 Ops/s $\color{#d91a1a}-2.84\%$
test_serial 1.3773s 1.3745s 0.7276 Ops/s 0.7175 Ops/s $\color{#35bf28}+1.40\%$
test_parallel 1.2195s 1.2056s 0.8295 Ops/s 0.8146 Ops/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[True-True-True-True-True] 0.6252ms 30.4393μs 32.8523 KOps/s 33.1706 KOps/s $\color{#d91a1a}-0.96\%$
test_step_mdp_speed[True-True-True-True-False] 41.5170μs 17.7510μs 56.3348 KOps/s 54.9216 KOps/s $\color{#35bf28}+2.57\%$
test_step_mdp_speed[True-True-True-False-True] 45.4250μs 16.9918μs 58.8518 KOps/s 57.9815 KOps/s $\color{#35bf28}+1.50\%$
test_step_mdp_speed[True-True-True-False-False] 36.3980μs 9.9744μs 100.2566 KOps/s 98.2723 KOps/s $\color{#35bf28}+2.02\%$
test_step_mdp_speed[True-True-False-True-True] 79.2280μs 32.1397μs 31.1142 KOps/s 31.0014 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[True-True-False-True-False] 44.9840μs 19.8143μs 50.4685 KOps/s 50.1734 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[True-True-False-False-True] 50.0730μs 18.9066μs 52.8916 KOps/s 52.6569 KOps/s $\color{#35bf28}+0.45\%$
test_step_mdp_speed[True-True-False-False-False] 47.0970μs 11.8377μs 84.4756 KOps/s 83.3799 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[True-False-True-True-True] 64.1600μs 34.0088μs 29.4041 KOps/s 29.4427 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-False-True-True-False] 47.9700μs 21.7001μs 46.0827 KOps/s 46.0958 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[True-False-True-False-True] 61.1430μs 18.9885μs 52.6635 KOps/s 52.2643 KOps/s $\color{#35bf28}+0.76\%$
test_step_mdp_speed[True-False-True-False-False] 37.3890μs 11.8391μs 84.4661 KOps/s 84.0889 KOps/s $\color{#35bf28}+0.45\%$
test_step_mdp_speed[True-False-False-True-True] 77.5450μs 35.7879μs 27.9424 KOps/s 27.5647 KOps/s $\color{#35bf28}+1.37\%$
test_step_mdp_speed[True-False-False-True-False] 51.1860μs 22.9811μs 43.5140 KOps/s 42.5641 KOps/s $\color{#35bf28}+2.23\%$
test_step_mdp_speed[True-False-False-False-True] 70.8420μs 20.5864μs 48.5758 KOps/s 48.2451 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[True-False-False-False-False] 40.4260μs 13.5884μs 73.5922 KOps/s 72.6737 KOps/s $\color{#35bf28}+1.26\%$
test_step_mdp_speed[False-True-True-True-True] 64.6110μs 33.9420μs 29.4621 KOps/s 29.3558 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[False-True-True-True-False] 57.3970μs 21.3990μs 46.7312 KOps/s 46.1503 KOps/s $\color{#35bf28}+1.26\%$
test_step_mdp_speed[False-True-True-False-True] 2.4405ms 21.6068μs 46.2817 KOps/s 46.3209 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[False-True-True-False-False] 38.6220μs 13.2514μs 75.4637 KOps/s 74.8730 KOps/s $\color{#35bf28}+0.79\%$
test_step_mdp_speed[False-True-False-True-True] 80.2400μs 35.3477μs 28.2904 KOps/s 27.7550 KOps/s $\color{#35bf28}+1.93\%$
test_step_mdp_speed[False-True-False-True-False] 54.0210μs 23.2289μs 43.0498 KOps/s 42.6787 KOps/s $\color{#35bf28}+0.87\%$
test_step_mdp_speed[False-True-False-False-True] 47.2380μs 23.3245μs 42.8734 KOps/s 42.3891 KOps/s $\color{#35bf28}+1.14\%$
test_step_mdp_speed[False-True-False-False-False] 41.0760μs 15.0934μs 66.2542 KOps/s 65.8894 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-False-True-True-True] 78.0660μs 38.3455μs 26.0787 KOps/s 26.4176 KOps/s $\color{#d91a1a}-1.28\%$
test_step_mdp_speed[False-False-True-True-False] 58.3980μs 25.1302μs 39.7927 KOps/s 38.7187 KOps/s $\color{#35bf28}+2.77\%$
test_step_mdp_speed[False-False-True-False-True] 62.1360μs 23.5257μs 42.5068 KOps/s 42.4784 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[False-False-True-False-False] 39.4430μs 15.0233μs 66.5632 KOps/s 63.9718 KOps/s $\color{#35bf28}+4.05\%$
test_step_mdp_speed[False-False-False-True-True] 72.6250μs 39.3117μs 25.4377 KOps/s 25.4700 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[False-False-False-True-False] 55.6840μs 26.9665μs 37.0831 KOps/s 37.3623 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-False-False-False-True] 52.8480μs 24.8660μs 40.2156 KOps/s 39.4409 KOps/s $\color{#35bf28}+1.96\%$
test_step_mdp_speed[False-False-False-False-False] 55.8040μs 16.5688μs 60.3544 KOps/s 59.7199 KOps/s $\color{#35bf28}+1.06\%$
test_values[generalized_advantage_estimate-True-True] 10.3747ms 9.8969ms 101.0419 Ops/s 104.7792 Ops/s $\color{#d91a1a}-3.57\%$
test_values[vec_generalized_advantage_estimate-True-True] 26.8289ms 24.4722ms 40.8627 Ops/s 40.8886 Ops/s $\color{#d91a1a}-0.06\%$
test_values[td0_return_estimate-False-False] 0.2491ms 0.1786ms 5.5988 KOps/s 5.5633 KOps/s $\color{#35bf28}+0.64\%$
test_values[td1_return_estimate-False-False] 29.2481ms 25.0721ms 39.8850 Ops/s 41.3760 Ops/s $\color{#d91a1a}-3.60\%$
test_values[vec_td1_return_estimate-False-False] 30.9490ms 24.8780ms 40.1961 Ops/s 40.9200 Ops/s $\color{#d91a1a}-1.77\%$
test_values[td_lambda_return_estimate-True-False] 38.1230ms 35.7079ms 28.0050 Ops/s 28.8733 Ops/s $\color{#d91a1a}-3.01\%$
test_values[vec_td_lambda_return_estimate-True-False] 26.6058ms 24.4836ms 40.8436 Ops/s 40.8833 Ops/s $\color{#d91a1a}-0.10\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.7591ms 8.6158ms 116.0660 Ops/s 120.7788 Ops/s $\color{#d91a1a}-3.90\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.6467ms 2.0235ms 494.2044 Ops/s 524.1325 Ops/s $\textbf{\color{#d91a1a}-5.71\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5151ms 0.3765ms 2.6559 KOps/s 2.7229 KOps/s $\color{#d91a1a}-2.46\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 49.2718ms 42.0236ms 23.7962 Ops/s 23.1934 Ops/s $\color{#35bf28}+2.60\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.7788ms 3.4576ms 289.2212 Ops/s 290.3251 Ops/s $\color{#d91a1a}-0.38\%$
test_dqn_speed[False-None] 1.7192ms 1.4368ms 695.9916 Ops/s 694.7188 Ops/s $\color{#35bf28}+0.18\%$
test_dqn_speed[False-backward] 2.3112ms 1.9273ms 518.8500 Ops/s 523.2298 Ops/s $\color{#d91a1a}-0.84\%$
test_dqn_speed[True-None] 0.6027ms 0.4858ms 2.0584 KOps/s 2.0087 KOps/s $\color{#35bf28}+2.47\%$
test_dqn_speed[True-backward] 1.0536ms 0.9574ms 1.0445 KOps/s 1.0361 KOps/s $\color{#35bf28}+0.81\%$
test_dqn_speed[reduce-overhead-None] 1.1574ms 0.4849ms 2.0621 KOps/s 2.0422 KOps/s $\color{#35bf28}+0.98\%$
test_dqn_speed[reduce-overhead-backward] 1.0643ms 0.9648ms 1.0365 KOps/s 1.0467 KOps/s $\color{#d91a1a}-0.97\%$
test_ddpg_speed[False-None] 3.9304ms 2.9677ms 336.9614 Ops/s 339.2877 Ops/s $\color{#d91a1a}-0.69\%$
test_ddpg_speed[False-backward] 4.5113ms 4.1134ms 243.1077 Ops/s 242.6145 Ops/s $\color{#35bf28}+0.20\%$
test_ddpg_speed[True-None] 1.9181ms 1.2669ms 789.3406 Ops/s 807.7206 Ops/s $\color{#d91a1a}-2.28\%$
test_ddpg_speed[True-backward] 2.1965ms 2.1250ms 470.5961 Ops/s 457.0174 Ops/s $\color{#35bf28}+2.97\%$
test_ddpg_speed[reduce-overhead-None] 1.7430ms 1.2240ms 817.0126 Ops/s 811.1470 Ops/s $\color{#35bf28}+0.72\%$
test_ddpg_speed[reduce-overhead-backward] 2.3477ms 2.1859ms 457.4841 Ops/s 471.2744 Ops/s $\color{#d91a1a}-2.93\%$
test_sac_speed[False-None] 9.6278ms 8.2399ms 121.3611 Ops/s 122.2697 Ops/s $\color{#d91a1a}-0.74\%$
test_sac_speed[False-backward] 12.8470ms 11.0601ms 90.4151 Ops/s 92.2114 Ops/s $\color{#d91a1a}-1.95\%$
test_sac_speed[True-None] 2.8084ms 2.0896ms 478.5494 Ops/s 464.8633 Ops/s $\color{#35bf28}+2.94\%$
test_sac_speed[True-backward] 4.0409ms 3.8586ms 259.1637 Ops/s 257.1084 Ops/s $\color{#35bf28}+0.80\%$
test_sac_speed[reduce-overhead-None] 2.6075ms 2.0863ms 479.3090 Ops/s 458.9933 Ops/s $\color{#35bf28}+4.43\%$
test_sac_speed[reduce-overhead-backward] 4.4019ms 3.7824ms 264.3797 Ops/s 260.3221 Ops/s $\color{#35bf28}+1.56\%$
test_redq_speed[False-None] 14.3997ms 13.2825ms 75.2871 Ops/s 74.8243 Ops/s $\color{#35bf28}+0.62\%$
test_redq_speed[False-backward] 24.4051ms 22.6391ms 44.1714 Ops/s 44.3222 Ops/s $\color{#d91a1a}-0.34\%$
test_redq_speed[True-None] 6.0422ms 5.0620ms 197.5514 Ops/s 188.1270 Ops/s $\textbf{\color{#35bf28}+5.01\%}$
test_redq_speed[True-backward] 13.9520ms 12.9994ms 76.9265 Ops/s 70.5051 Ops/s $\textbf{\color{#35bf28}+9.11\%}$
test_redq_speed[reduce-overhead-None] 6.4275ms 5.1807ms 193.0233 Ops/s 173.6132 Ops/s $\textbf{\color{#35bf28}+11.18\%}$
test_redq_speed[reduce-overhead-backward] 14.7120ms 12.8280ms 77.9544 Ops/s 74.4800 Ops/s $\color{#35bf28}+4.66\%$
test_redq_deprec_speed[False-None] 15.0769ms 13.3296ms 75.0212 Ops/s 74.4146 Ops/s $\color{#35bf28}+0.82\%$
test_redq_deprec_speed[False-backward] 19.8644ms 19.0094ms 52.6055 Ops/s 51.3660 Ops/s $\color{#35bf28}+2.41\%$
test_redq_deprec_speed[True-None] 4.3113ms 3.8650ms 258.7329 Ops/s 246.6877 Ops/s $\color{#35bf28}+4.88\%$
test_redq_deprec_speed[True-backward] 9.6656ms 8.5492ms 116.9704 Ops/s 114.2045 Ops/s $\color{#35bf28}+2.42\%$
test_redq_deprec_speed[reduce-overhead-None] 4.5692ms 3.8473ms 259.9241 Ops/s 246.9860 Ops/s $\textbf{\color{#35bf28}+5.24\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.6670ms 8.6785ms 115.2275 Ops/s 115.1934 Ops/s $\color{#35bf28}+0.03\%$
test_td3_speed[False-None] 8.8398ms 8.1841ms 122.1878 Ops/s 120.7691 Ops/s $\color{#35bf28}+1.17\%$
test_td3_speed[False-backward] 11.6802ms 10.7636ms 92.9060 Ops/s 93.3936 Ops/s $\color{#d91a1a}-0.52\%$
test_td3_speed[True-None] 1.8428ms 1.7690ms 565.2933 Ops/s 548.7368 Ops/s $\color{#35bf28}+3.02\%$
test_td3_speed[True-backward] 3.5285ms 3.3790ms 295.9454 Ops/s 288.7216 Ops/s $\color{#35bf28}+2.50\%$
test_td3_speed[reduce-overhead-None] 1.9633ms 1.7668ms 566.0053 Ops/s 549.1104 Ops/s $\color{#35bf28}+3.08\%$
test_td3_speed[reduce-overhead-backward] 3.5347ms 3.3635ms 297.3101 Ops/s 287.9978 Ops/s $\color{#35bf28}+3.23\%$
test_cql_speed[False-None] 40.3018ms 36.6765ms 27.2654 Ops/s 26.9850 Ops/s $\color{#35bf28}+1.04\%$
test_cql_speed[False-backward] 59.8430ms 47.8242ms 20.9099 Ops/s 20.9828 Ops/s $\color{#d91a1a}-0.35\%$
test_cql_speed[True-None] 17.4873ms 16.1640ms 61.8658 Ops/s 60.9763 Ops/s $\color{#35bf28}+1.46\%$
test_cql_speed[True-backward] 24.1776ms 23.1555ms 43.1863 Ops/s 42.3592 Ops/s $\color{#35bf28}+1.95\%$
test_cql_speed[reduce-overhead-None] 17.5693ms 16.3058ms 61.3279 Ops/s 60.8820 Ops/s $\color{#35bf28}+0.73\%$
test_cql_speed[reduce-overhead-backward] 24.1796ms 23.4416ms 42.6593 Ops/s 42.8254 Ops/s $\color{#d91a1a}-0.39\%$
test_a2c_speed[False-None] 8.6104ms 7.3343ms 136.3448 Ops/s 136.5947 Ops/s $\color{#d91a1a}-0.18\%$
test_a2c_speed[False-backward] 15.9541ms 14.6175ms 68.4113 Ops/s 67.7647 Ops/s $\color{#35bf28}+0.95\%$
test_a2c_speed[True-None] 4.1040ms 3.7044ms 269.9495 Ops/s 267.0607 Ops/s $\color{#35bf28}+1.08\%$
test_a2c_speed[True-backward] 10.7672ms 10.2342ms 97.7114 Ops/s 95.2308 Ops/s $\color{#35bf28}+2.60\%$
test_a2c_speed[reduce-overhead-None] 4.3600ms 3.7267ms 268.3345 Ops/s 263.6015 Ops/s $\color{#35bf28}+1.80\%$
test_a2c_speed[reduce-overhead-backward] 10.8514ms 10.3719ms 96.4145 Ops/s 95.2345 Ops/s $\color{#35bf28}+1.24\%$
test_ppo_speed[False-None] 11.2185ms 7.7928ms 128.3241 Ops/s 130.8454 Ops/s $\color{#d91a1a}-1.93\%$
test_ppo_speed[False-backward] 15.7256ms 15.1123ms 66.1715 Ops/s 66.5135 Ops/s $\color{#d91a1a}-0.51\%$
test_ppo_speed[True-None] 5.1646ms 4.1135ms 243.1032 Ops/s 240.9269 Ops/s $\color{#35bf28}+0.90\%$
test_ppo_speed[True-backward] 11.1207ms 10.2340ms 97.7134 Ops/s 96.2953 Ops/s $\color{#35bf28}+1.47\%$
test_ppo_speed[reduce-overhead-None] 4.3309ms 4.0812ms 245.0244 Ops/s 240.0444 Ops/s $\color{#35bf28}+2.07\%$
test_ppo_speed[reduce-overhead-backward] 11.0660ms 10.1725ms 98.3040 Ops/s 94.5140 Ops/s $\color{#35bf28}+4.01\%$
test_reinforce_speed[False-None] 7.8236ms 6.6101ms 151.2837 Ops/s 149.3732 Ops/s $\color{#35bf28}+1.28\%$
test_reinforce_speed[False-backward] 10.4060ms 9.9477ms 100.5262 Ops/s 99.7018 Ops/s $\color{#35bf28}+0.83\%$
test_reinforce_speed[True-None] 3.4100ms 3.0488ms 327.9969 Ops/s 320.1800 Ops/s $\color{#35bf28}+2.44\%$
test_reinforce_speed[True-backward] 9.7478ms 9.1575ms 109.2003 Ops/s 107.2747 Ops/s $\color{#35bf28}+1.80\%$
test_reinforce_speed[reduce-overhead-None] 3.6106ms 3.0686ms 325.8850 Ops/s 313.1172 Ops/s $\color{#35bf28}+4.08\%$
test_reinforce_speed[reduce-overhead-backward] 9.5985ms 9.1619ms 109.1477 Ops/s 107.5052 Ops/s $\color{#35bf28}+1.53\%$
test_iql_speed[False-None] 39.3946ms 33.0157ms 30.2886 Ops/s 29.8668 Ops/s $\color{#35bf28}+1.41\%$
test_iql_speed[False-backward] 46.7637ms 45.5775ms 21.9407 Ops/s 21.5672 Ops/s $\color{#35bf28}+1.73\%$
test_iql_speed[True-None] 12.6484ms 11.5454ms 86.6146 Ops/s 86.8645 Ops/s $\color{#d91a1a}-0.29\%$
test_iql_speed[True-backward] 24.6615ms 22.7936ms 43.8719 Ops/s 43.5913 Ops/s $\color{#35bf28}+0.64\%$
test_iql_speed[reduce-overhead-None] 13.2699ms 11.5988ms 86.2160 Ops/s 83.6884 Ops/s $\color{#35bf28}+3.02\%$
test_iql_speed[reduce-overhead-backward] 23.4620ms 22.6862ms 44.0797 Ops/s 43.0275 Ops/s $\color{#35bf28}+2.45\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.5153ms 5.1205ms 195.2926 Ops/s 194.3142 Ops/s $\color{#35bf28}+0.50\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7629ms 0.5430ms 1.8415 KOps/s 1.7131 KOps/s $\textbf{\color{#35bf28}+7.50\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8097ms 0.5200ms 1.9232 KOps/s 1.9034 KOps/s $\color{#35bf28}+1.04\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.0817ms 4.7992ms 208.3690 Ops/s 205.2459 Ops/s $\color{#35bf28}+1.52\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8758ms 0.5326ms 1.8777 KOps/s 1.8405 KOps/s $\color{#35bf28}+2.02\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8380ms 0.5143ms 1.9446 KOps/s 1.9024 KOps/s $\color{#35bf28}+2.22\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.4336ms 1.7443ms 573.2872 Ops/s 564.8157 Ops/s $\color{#35bf28}+1.50\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.0100ms 1.6517ms 605.4318 Ops/s 593.7703 Ops/s $\color{#35bf28}+1.96\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.1876ms 4.9329ms 202.7211 Ops/s 199.1030 Ops/s $\color{#35bf28}+1.82\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.3003ms 0.6894ms 1.4505 KOps/s 1.4197 KOps/s $\color{#35bf28}+2.17\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9529ms 0.6561ms 1.5241 KOps/s 1.5014 KOps/s $\color{#35bf28}+1.51\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.3889ms 4.8420ms 206.5269 Ops/s 204.8954 Ops/s $\color{#35bf28}+0.80\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9207ms 0.5544ms 1.8036 KOps/s 1.7297 KOps/s $\color{#35bf28}+4.27\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8386ms 0.5230ms 1.9122 KOps/s 1.8970 KOps/s $\color{#35bf28}+0.80\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.4187ms 4.8370ms 206.7413 Ops/s 207.6987 Ops/s $\color{#d91a1a}-0.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9546ms 0.5391ms 1.8551 KOps/s 1.7942 KOps/s $\color{#35bf28}+3.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 1.0088ms 0.5219ms 1.9161 KOps/s 1.8859 KOps/s $\color{#35bf28}+1.60\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.2061ms 5.0589ms 197.6721 Ops/s 198.9683 Ops/s $\color{#d91a1a}-0.65\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0965ms 0.7074ms 1.4136 KOps/s 1.4484 KOps/s $\color{#d91a1a}-2.40\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9312ms 0.6781ms 1.4747 KOps/s 1.4596 KOps/s $\color{#35bf28}+1.03\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.3996ms 4.1595ms 240.4144 Ops/s 240.8687 Ops/s $\color{#d91a1a}-0.19\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.2510ms 2.4752ms 404.0075 Ops/s 398.3679 Ops/s $\color{#35bf28}+1.42\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.6978ms 1.3617ms 734.3928 Ops/s 701.1368 Ops/s $\color{#35bf28}+4.74\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.3304ms 4.2993ms 232.5953 Ops/s 247.8124 Ops/s $\textbf{\color{#d91a1a}-6.14\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.2605ms 2.3904ms 418.3355 Ops/s 397.3955 Ops/s $\textbf{\color{#35bf28}+5.27\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.5406ms 1.4689ms 680.7652 Ops/s 751.0724 Ops/s $\textbf{\color{#d91a1a}-9.36\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4592s 13.5201ms 73.9642 Ops/s 34.1069 Ops/s $\textbf{\color{#35bf28}+116.86\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.7569ms 2.6637ms 375.4149 Ops/s 377.2305 Ops/s $\color{#d91a1a}-0.48\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.2144ms 1.6057ms 622.8004 Ops/s 656.3821 Ops/s $\textbf{\color{#d91a1a}-5.12\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 12.8337ms 11.5579ms 86.5208 Ops/s 82.1121 Ops/s $\textbf{\color{#35bf28}+5.37\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.4163ms 14.5363ms 68.7931 Ops/s 67.3130 Ops/s $\color{#35bf28}+2.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 21.1490ms 20.2422ms 49.4019 Ops/s 47.7488 Ops/s $\color{#35bf28}+3.46\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.3598ms 14.6220ms 68.3902 Ops/s 65.6638 Ops/s $\color{#35bf28}+4.15\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.3235ms 20.3146ms 49.2257 Ops/s 47.4966 Ops/s $\color{#35bf28}+3.64\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 16.7947ms 16.0666ms 62.2407 Ops/s 60.5875 Ops/s $\color{#35bf28}+2.73\%$

Copy link

github-actions bot commented Feb 4, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8270s 0.7418s 1.3480 Ops/s 1.3858 Ops/s $\color{#d91a1a}-2.73\%$
test_transformed 1.4053s 1.3179s 0.7588 Ops/s 0.7536 Ops/s $\color{#35bf28}+0.68\%$
test_serial 2.1443s 2.1325s 0.4689 Ops/s 0.4612 Ops/s $\color{#35bf28}+1.69\%$
test_parallel 1.8282s 1.8182s 0.5500 Ops/s 0.5432 Ops/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-True-True-True-True] 0.2176ms 40.3938μs 24.7563 KOps/s 24.5806 KOps/s $\color{#35bf28}+0.71\%$
test_step_mdp_speed[True-True-True-True-False] 52.9110μs 23.3374μs 42.8496 KOps/s 42.5786 KOps/s $\color{#35bf28}+0.64\%$
test_step_mdp_speed[True-True-True-False-True] 53.1120μs 21.8665μs 45.7321 KOps/s 44.8129 KOps/s $\color{#35bf28}+2.05\%$
test_step_mdp_speed[True-True-True-False-False] 68.7410μs 12.3354μs 81.0674 KOps/s 77.4775 KOps/s $\color{#35bf28}+4.63\%$
test_step_mdp_speed[True-True-False-True-True] 83.4420μs 42.1067μs 23.7492 KOps/s 23.2365 KOps/s $\color{#35bf28}+2.21\%$
test_step_mdp_speed[True-True-False-True-False] 55.9110μs 25.0431μs 39.9311 KOps/s 38.7670 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[True-True-False-False-True] 56.5910μs 24.1511μs 41.4061 KOps/s 40.8359 KOps/s $\color{#35bf28}+1.40\%$
test_step_mdp_speed[True-True-False-False-False] 42.6300μs 15.1574μs 65.9742 KOps/s 64.6903 KOps/s $\color{#35bf28}+1.98\%$
test_step_mdp_speed[True-False-True-True-True] 81.8320μs 44.9480μs 22.2479 KOps/s 23.1644 KOps/s $\color{#d91a1a}-3.96\%$
test_step_mdp_speed[True-False-True-True-False] 76.6920μs 28.2694μs 35.3740 KOps/s 35.4799 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-False-True-False-True] 54.6310μs 24.6040μs 40.6438 KOps/s 41.1400 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[True-False-True-False-False] 50.9710μs 15.3031μs 65.3462 KOps/s 65.4258 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-False-False-True-True] 82.2510μs 46.6108μs 21.4543 KOps/s 21.2143 KOps/s $\color{#35bf28}+1.13\%$
test_step_mdp_speed[True-False-False-True-False] 56.1910μs 30.4722μs 32.8168 KOps/s 33.0752 KOps/s $\color{#d91a1a}-0.78\%$
test_step_mdp_speed[True-False-False-False-True] 60.8420μs 26.8568μs 37.2346 KOps/s 37.3736 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-False-False-False-False] 43.2510μs 17.5572μs 56.9568 KOps/s 56.7785 KOps/s $\color{#35bf28}+0.31\%$
test_step_mdp_speed[False-True-True-True-True] 81.0610μs 45.1952μs 22.1262 KOps/s 22.4156 KOps/s $\color{#d91a1a}-1.29\%$
test_step_mdp_speed[False-True-True-True-False] 61.7720μs 28.1752μs 35.4923 KOps/s 35.2027 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[False-True-True-False-True] 2.6232ms 28.6267μs 34.9325 KOps/s 35.2151 KOps/s $\color{#d91a1a}-0.80\%$
test_step_mdp_speed[False-True-True-False-False] 45.6010μs 17.1564μs 58.2874 KOps/s 58.1913 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[False-True-False-True-True] 80.2920μs 47.2192μs 21.1778 KOps/s 21.3297 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[False-True-False-True-False] 59.3810μs 30.4434μs 32.8479 KOps/s 32.7552 KOps/s $\color{#35bf28}+0.28\%$
test_step_mdp_speed[False-True-False-False-True] 58.7310μs 30.4400μs 32.8515 KOps/s 31.7457 KOps/s $\color{#35bf28}+3.48\%$
test_step_mdp_speed[False-True-False-False-False] 48.1510μs 19.4288μs 51.4701 KOps/s 51.0184 KOps/s $\color{#35bf28}+0.89\%$
test_step_mdp_speed[False-False-True-True-True] 78.6710μs 50.0435μs 19.9826 KOps/s 20.1870 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-False-True-True-False] 61.0410μs 32.6378μs 30.6393 KOps/s 30.5145 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[False-False-True-False-True] 61.0010μs 30.4443μs 32.8469 KOps/s 32.2812 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[False-False-True-False-False] 49.9110μs 18.8839μs 52.9551 KOps/s 51.9769 KOps/s $\color{#35bf28}+1.88\%$
test_step_mdp_speed[False-False-False-True-True] 93.8010μs 50.2694μs 19.8928 KOps/s 19.4864 KOps/s $\color{#35bf28}+2.09\%$
test_step_mdp_speed[False-False-False-True-False] 65.5810μs 35.1161μs 28.4770 KOps/s 28.8386 KOps/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[False-False-False-False-True] 59.9910μs 32.2046μs 31.0515 KOps/s 31.3707 KOps/s $\color{#d91a1a}-1.02\%$
test_step_mdp_speed[False-False-False-False-False] 47.8810μs 20.9804μs 47.6636 KOps/s 46.3635 KOps/s $\color{#35bf28}+2.80\%$
test_values[generalized_advantage_estimate-True-True] 26.1182ms 24.9990ms 40.0016 Ops/s 39.2563 Ops/s $\color{#35bf28}+1.90\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1158s 3.2162ms 310.9271 Ops/s 332.5211 Ops/s $\textbf{\color{#d91a1a}-6.49\%}$
test_values[td0_return_estimate-False-False] 0.1137ms 78.8843μs 12.6768 KOps/s 12.2388 KOps/s $\color{#35bf28}+3.58\%$
test_values[td1_return_estimate-False-False] 58.9970ms 56.3725ms 17.7392 Ops/s 17.3571 Ops/s $\color{#35bf28}+2.20\%$
test_values[vec_td1_return_estimate-False-False] 1.2891ms 1.0962ms 912.2234 Ops/s 917.3287 Ops/s $\color{#d91a1a}-0.56\%$
test_values[td_lambda_return_estimate-True-False] 93.7107ms 89.5323ms 11.1692 Ops/s 10.8412 Ops/s $\color{#35bf28}+3.02\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.4249ms 1.0888ms 918.4451 Ops/s 923.3330 Ops/s $\color{#d91a1a}-0.53\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.4488ms 24.9022ms 40.1571 Ops/s 38.2543 Ops/s $\color{#35bf28}+4.97\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0458ms 0.7570ms 1.3210 KOps/s 1.3226 KOps/s $\color{#d91a1a}-0.12\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7711ms 0.6715ms 1.4891 KOps/s 1.4912 KOps/s $\color{#d91a1a}-0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5199ms 1.4843ms 673.6994 Ops/s 670.4321 Ops/s $\color{#35bf28}+0.49\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8096ms 0.6923ms 1.4446 KOps/s 1.4408 KOps/s $\color{#35bf28}+0.26\%$
test_dqn_speed[False-None] 7.0340ms 1.5197ms 658.0104 Ops/s 642.0999 Ops/s $\color{#35bf28}+2.48\%$
test_dqn_speed[False-backward] 2.3941ms 2.1268ms 470.2005 Ops/s 460.7356 Ops/s $\color{#35bf28}+2.05\%$
test_dqn_speed[True-None] 0.6464ms 0.5653ms 1.7691 KOps/s 1.7683 KOps/s $\color{#35bf28}+0.04\%$
test_dqn_speed[True-backward] 1.3019ms 1.2321ms 811.6065 Ops/s 877.4388 Ops/s $\textbf{\color{#d91a1a}-7.50\%}$
test_dqn_speed[reduce-overhead-None] 0.6881ms 0.5875ms 1.7022 KOps/s 1.7291 KOps/s $\color{#d91a1a}-1.55\%$
test_dqn_speed[reduce-overhead-backward] 1.1171ms 1.0719ms 932.9183 Ops/s 1.0267 KOps/s $\textbf{\color{#d91a1a}-9.14\%}$
test_ddpg_speed[False-None] 3.2487ms 2.9000ms 344.8289 Ops/s 342.2694 Ops/s $\color{#35bf28}+0.75\%$
test_ddpg_speed[False-backward] 4.6895ms 4.2838ms 233.4362 Ops/s 239.4175 Ops/s $\color{#d91a1a}-2.50\%$
test_ddpg_speed[True-None] 1.4257ms 1.3388ms 746.9303 Ops/s 737.6791 Ops/s $\color{#35bf28}+1.25\%$
test_ddpg_speed[True-backward] 2.6214ms 2.5644ms 389.9570 Ops/s 405.9168 Ops/s $\color{#d91a1a}-3.93\%$
test_ddpg_speed[reduce-overhead-None] 1.4913ms 1.3478ms 741.9637 Ops/s 730.3943 Ops/s $\color{#35bf28}+1.58\%$
test_ddpg_speed[reduce-overhead-backward] 2.1090ms 2.0138ms 496.5853 Ops/s 527.4737 Ops/s $\textbf{\color{#d91a1a}-5.86\%}$
test_sac_speed[False-None] 8.4122ms 7.9848ms 125.2381 Ops/s 123.1664 Ops/s $\color{#35bf28}+1.68\%$
test_sac_speed[False-backward] 11.5820ms 11.1398ms 89.7682 Ops/s 90.6140 Ops/s $\color{#d91a1a}-0.93\%$
test_sac_speed[True-None] 1.9051ms 1.8356ms 544.7953 Ops/s 534.9054 Ops/s $\color{#35bf28}+1.85\%$
test_sac_speed[True-backward] 4.1853ms 3.7212ms 268.7327 Ops/s 274.9783 Ops/s $\color{#d91a1a}-2.27\%$
test_sac_speed[reduce-overhead-None] 21.6764ms 12.0600ms 82.9184 Ops/s 82.6561 Ops/s $\color{#35bf28}+0.32\%$
test_sac_speed[reduce-overhead-backward] 1.8799ms 1.7929ms 557.7517 Ops/s 592.9403 Ops/s $\textbf{\color{#d91a1a}-5.93\%}$
test_redq_speed[False-None] 7.9493ms 7.4005ms 135.1266 Ops/s 130.8703 Ops/s $\color{#35bf28}+3.25\%$
test_redq_speed[False-backward] 12.0342ms 11.5752ms 86.3918 Ops/s 86.6988 Ops/s $\color{#d91a1a}-0.35\%$
test_redq_speed[True-None] 2.3902ms 2.3030ms 434.2209 Ops/s 425.2158 Ops/s $\color{#35bf28}+2.12\%$
test_redq_speed[True-backward] 4.5812ms 4.1733ms 239.6164 Ops/s 243.0688 Ops/s $\color{#d91a1a}-1.42\%$
test_redq_speed[reduce-overhead-None] 2.5393ms 2.3202ms 431.0027 Ops/s 419.9280 Ops/s $\color{#35bf28}+2.64\%$
test_redq_speed[reduce-overhead-backward] 4.6151ms 4.1950ms 238.3814 Ops/s 237.5110 Ops/s $\color{#35bf28}+0.37\%$
test_redq_deprec_speed[False-None] 9.4105ms 9.0329ms 110.7062 Ops/s 108.5813 Ops/s $\color{#35bf28}+1.96\%$
test_redq_deprec_speed[False-backward] 12.7979ms 12.3071ms 81.2537 Ops/s 82.1203 Ops/s $\color{#d91a1a}-1.06\%$
test_redq_deprec_speed[True-None] 2.7351ms 2.6412ms 378.6153 Ops/s 370.1903 Ops/s $\color{#35bf28}+2.28\%$
test_redq_deprec_speed[True-backward] 4.8601ms 4.4510ms 224.6688 Ops/s 218.3172 Ops/s $\color{#35bf28}+2.91\%$
test_redq_deprec_speed[reduce-overhead-None] 2.7551ms 2.6255ms 380.8737 Ops/s 368.7996 Ops/s $\color{#35bf28}+3.27\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.9909ms 4.5364ms 220.4412 Ops/s 221.0909 Ops/s $\color{#d91a1a}-0.29\%$
test_td3_speed[False-None] 8.0042ms 7.9339ms 126.0415 Ops/s 123.8381 Ops/s $\color{#35bf28}+1.78\%$
test_td3_speed[False-backward] 11.2891ms 10.5599ms 94.6978 Ops/s 95.0773 Ops/s $\color{#d91a1a}-0.40\%$
test_td3_speed[True-None] 1.7538ms 1.6566ms 603.6344 Ops/s 588.9713 Ops/s $\color{#35bf28}+2.49\%$
test_td3_speed[True-backward] 3.9104ms 3.3663ms 297.0605 Ops/s 307.2414 Ops/s $\color{#d91a1a}-3.31\%$
test_td3_speed[reduce-overhead-None] 55.2575ms 26.5647ms 37.6439 Ops/s 36.1714 Ops/s $\color{#35bf28}+4.07\%$
test_td3_speed[reduce-overhead-backward] 1.5669ms 1.4968ms 668.0917 Ops/s 706.4530 Ops/s $\textbf{\color{#d91a1a}-5.43\%}$
test_cql_speed[False-None] 17.2513ms 16.7194ms 59.8107 Ops/s 58.8234 Ops/s $\color{#35bf28}+1.68\%$
test_cql_speed[False-backward] 22.6959ms 22.1891ms 45.0671 Ops/s 45.0446 Ops/s $\color{#35bf28}+0.05\%$
test_cql_speed[True-None] 3.3559ms 3.2626ms 306.5059 Ops/s 297.9220 Ops/s $\color{#35bf28}+2.88\%$
test_cql_speed[True-backward] 6.0942ms 5.6065ms 178.3643 Ops/s 171.1348 Ops/s $\color{#35bf28}+4.22\%$
test_cql_speed[reduce-overhead-None] 21.3426ms 13.2721ms 75.3461 Ops/s 57.5420 Ops/s $\textbf{\color{#35bf28}+30.94\%}$
test_cql_speed[reduce-overhead-backward] 1.9630ms 1.8160ms 550.6743 Ops/s 488.6871 Ops/s $\textbf{\color{#35bf28}+12.68\%}$
test_a2c_speed[False-None] 3.2947ms 3.1555ms 316.9057 Ops/s 306.2962 Ops/s $\color{#35bf28}+3.46\%$
test_a2c_speed[False-backward] 6.6968ms 6.1135ms 163.5723 Ops/s 157.1592 Ops/s $\color{#35bf28}+4.08\%$
test_a2c_speed[True-None] 1.7794ms 1.3479ms 741.9113 Ops/s 736.6705 Ops/s $\color{#35bf28}+0.71\%$
test_a2c_speed[True-backward] 3.3883ms 2.9068ms 344.0229 Ops/s 318.0165 Ops/s $\textbf{\color{#35bf28}+8.18\%}$
test_a2c_speed[reduce-overhead-None] 16.1109ms 9.1456ms 109.3416 Ops/s 109.4923 Ops/s $\color{#d91a1a}-0.14\%$
test_a2c_speed[reduce-overhead-backward] 1.5461ms 1.4684ms 681.0206 Ops/s 673.9988 Ops/s $\color{#35bf28}+1.04\%$
test_ppo_speed[False-None] 3.7795ms 3.6769ms 271.9653 Ops/s 263.3290 Ops/s $\color{#35bf28}+3.28\%$
test_ppo_speed[False-backward] 7.4215ms 6.8990ms 144.9496 Ops/s 143.9446 Ops/s $\color{#35bf28}+0.70\%$
test_ppo_speed[True-None] 1.4792ms 1.4123ms 708.0699 Ops/s 698.1488 Ops/s $\color{#35bf28}+1.42\%$
test_ppo_speed[True-backward] 3.1263ms 3.0573ms 327.0885 Ops/s 304.7151 Ops/s $\textbf{\color{#35bf28}+7.34\%}$
test_ppo_speed[reduce-overhead-None] 1.0538ms 0.9748ms 1.0259 KOps/s 1.0267 KOps/s $\color{#d91a1a}-0.08\%$
test_ppo_speed[reduce-overhead-backward] 1.5128ms 1.4055ms 711.4968 Ops/s 617.5306 Ops/s $\textbf{\color{#35bf28}+15.22\%}$
test_reinforce_speed[False-None] 2.3980ms 2.2779ms 438.9992 Ops/s 432.0172 Ops/s $\color{#35bf28}+1.62\%$
test_reinforce_speed[False-backward] 3.7802ms 3.3531ms 298.2297 Ops/s 283.1475 Ops/s $\textbf{\color{#35bf28}+5.33\%}$
test_reinforce_speed[True-None] 1.3692ms 1.2928ms 773.5286 Ops/s 753.7097 Ops/s $\color{#35bf28}+2.63\%$
test_reinforce_speed[True-backward] 3.4458ms 2.9888ms 334.5808 Ops/s 319.4992 Ops/s $\color{#35bf28}+4.72\%$
test_reinforce_speed[reduce-overhead-None] 18.1436ms 10.0431ms 99.5707 Ops/s 99.9248 Ops/s $\color{#d91a1a}-0.35\%$
test_reinforce_speed[reduce-overhead-backward] 1.6113ms 1.5076ms 663.3186 Ops/s 594.2781 Ops/s $\textbf{\color{#35bf28}+11.62\%}$
test_iql_speed[False-None] 9.5921ms 9.1419ms 109.3861 Ops/s 105.4702 Ops/s $\color{#35bf28}+3.71\%$
test_iql_speed[False-backward] 13.3361ms 12.9063ms 77.4814 Ops/s 74.3218 Ops/s $\color{#35bf28}+4.25\%$
test_iql_speed[True-None] 2.3949ms 2.2423ms 445.9698 Ops/s 430.2179 Ops/s $\color{#35bf28}+3.66\%$
test_iql_speed[True-backward] 5.5747ms 4.9630ms 201.4890 Ops/s 199.9769 Ops/s $\color{#35bf28}+0.76\%$
test_iql_speed[reduce-overhead-None] 19.0240ms 11.2713ms 88.7207 Ops/s 90.0604 Ops/s $\color{#d91a1a}-1.49\%$
test_iql_speed[reduce-overhead-backward] 2.1403ms 2.0541ms 486.8251 Ops/s 484.4333 Ops/s $\color{#35bf28}+0.49\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9355ms 6.3667ms 157.0682 Ops/s 153.8230 Ops/s $\color{#35bf28}+2.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.4982ms 0.3156ms 3.1687 KOps/s 3.5743 KOps/s $\textbf{\color{#d91a1a}-11.35\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6695ms 0.3173ms 3.1521 KOps/s 3.4360 KOps/s $\textbf{\color{#d91a1a}-8.26\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.4066ms 6.0865ms 164.2973 Ops/s 162.6630 Ops/s $\color{#35bf28}+1.00\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6685ms 0.2684ms 3.7260 KOps/s 3.7835 KOps/s $\color{#d91a1a}-1.52\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6925ms 0.2470ms 4.0484 KOps/s 4.0476 KOps/s $\color{#35bf28}+0.02\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7414ms 1.2990ms 769.8186 Ops/s 772.1069 Ops/s $\color{#d91a1a}-0.30\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5394ms 1.3056ms 765.9179 Ops/s 739.4104 Ops/s $\color{#35bf28}+3.58\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6500ms 6.2551ms 159.8706 Ops/s 156.3248 Ops/s $\color{#35bf28}+2.27\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8990ms 0.4458ms 2.2430 KOps/s 2.4087 KOps/s $\textbf{\color{#d91a1a}-6.88\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8278ms 0.4323ms 2.3133 KOps/s 2.3580 KOps/s $\color{#d91a1a}-1.90\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.9546ms 6.1495ms 162.6139 Ops/s 157.6058 Ops/s $\color{#35bf28}+3.18\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.9191ms 0.3195ms 3.1303 KOps/s 3.1119 KOps/s $\color{#35bf28}+0.59\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8141ms 0.3208ms 3.1170 KOps/s 3.2084 KOps/s $\color{#d91a1a}-2.85\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.4238ms 6.0595ms 165.0307 Ops/s 160.0510 Ops/s $\color{#35bf28}+3.11\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7465ms 0.2794ms 3.5793 KOps/s 2.8969 KOps/s $\textbf{\color{#35bf28}+23.56\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5570ms 0.2716ms 3.6813 KOps/s 2.9739 KOps/s $\textbf{\color{#35bf28}+23.79\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.7451ms 6.2388ms 160.2867 Ops/s 154.6967 Ops/s $\color{#35bf28}+3.61\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.2481ms 0.4291ms 2.3307 KOps/s 2.1787 KOps/s $\textbf{\color{#35bf28}+6.98\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8330ms 0.3963ms 2.5230 KOps/s 2.5127 KOps/s $\color{#35bf28}+0.41\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1424ms 5.5295ms 180.8491 Ops/s 178.1695 Ops/s $\color{#35bf28}+1.50\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.8877ms 1.8451ms 541.9778 Ops/s 432.1869 Ops/s $\textbf{\color{#35bf28}+25.40\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.7308ms 1.3020ms 768.0732 Ops/s 756.2718 Ops/s $\color{#35bf28}+1.56\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 8.0942ms 5.6740ms 176.2417 Ops/s 176.1868 Ops/s $\color{#35bf28}+0.03\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.5492ms 2.0500ms 487.8083 Ops/s 411.0305 Ops/s $\textbf{\color{#35bf28}+18.68\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.4207ms 1.2458ms 802.7124 Ops/s 823.9582 Ops/s $\color{#d91a1a}-2.58\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4941s 15.6228ms 64.0089 Ops/s 31.5170 Ops/s $\textbf{\color{#35bf28}+103.09\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.6451ms 2.2511ms 444.2327 Ops/s 425.8970 Ops/s $\color{#35bf28}+4.31\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.9285ms 1.3947ms 716.9785 Ops/s 768.1238 Ops/s $\textbf{\color{#d91a1a}-6.66\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.5460ms 13.0383ms 76.6973 Ops/s 74.4292 Ops/s $\color{#35bf28}+3.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.2689ms 16.6668ms 59.9994 Ops/s 60.1215 Ops/s $\color{#d91a1a}-0.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.1797ms 17.6557ms 56.6388 Ops/s 55.4333 Ops/s $\color{#35bf28}+2.17\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.4091ms 16.8173ms 59.4627 Ops/s 57.5372 Ops/s $\color{#35bf28}+3.35\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.2340ms 17.6145ms 56.7713 Ops/s 55.3460 Ops/s $\color{#35bf28}+2.58\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.2566ms 18.1108ms 55.2158 Ops/s 55.0139 Ops/s $\color{#35bf28}+0.37\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants