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

[BugFix] Ensure that Composite.set returns self as TensorDict does #2784

Merged
merged 6 commits into from
Feb 13, 2025

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 12, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2784

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

❌ 2 New Failures, 2 Pending, 1 Unrelated Failure

As of commit 46c7263 with merge base f1c42e0 (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

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

@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 12, 2025
vmoens added a commit that referenced this pull request Feb 12, 2025
ghstack-source-id: ee47d30c335a95b5a100ba4a32f10c578cacdbdc
Pull Request resolved: #2784
Copy link

github-actions bot commented Feb 12, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}1$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5960s 0.5092s 1.9637 Ops/s 1.9607 Ops/s $\color{#35bf28}+0.15\%$
test_transformed 1.0861s 0.9938s 1.0062 Ops/s 1.0179 Ops/s $\color{#d91a1a}-1.14\%$
test_serial 1.5991s 1.5064s 0.6638 Ops/s 0.6654 Ops/s $\color{#d91a1a}-0.24\%$
test_parallel 1.3853s 1.2980s 0.7704 Ops/s 0.7715 Ops/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[True-True-True-True-True] 0.2165ms 30.0407μs 33.2882 KOps/s 33.7258 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[True-True-True-True-False] 47.9100μs 17.9136μs 55.8236 KOps/s 56.4158 KOps/s $\color{#d91a1a}-1.05\%$
test_step_mdp_speed[True-True-True-False-True] 0.6310ms 17.1492μs 58.3116 KOps/s 59.2758 KOps/s $\color{#d91a1a}-1.63\%$
test_step_mdp_speed[True-True-True-False-False] 46.1070μs 10.1362μs 98.6562 KOps/s 100.0174 KOps/s $\color{#d91a1a}-1.36\%$
test_step_mdp_speed[True-True-False-True-True] 67.2060μs 32.1140μs 31.1391 KOps/s 31.0019 KOps/s $\color{#35bf28}+0.44\%$
test_step_mdp_speed[True-True-False-True-False] 66.1010μs 19.6468μs 50.8990 KOps/s 50.4030 KOps/s $\color{#35bf28}+0.98\%$
test_step_mdp_speed[True-True-False-False-True] 71.5950μs 18.8785μs 52.9702 KOps/s 53.5520 KOps/s $\color{#d91a1a}-1.09\%$
test_step_mdp_speed[True-True-False-False-False] 39.5940μs 11.8779μs 84.1903 KOps/s 83.7837 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[True-False-True-True-True] 90.2790μs 34.0482μs 29.3701 KOps/s 29.6863 KOps/s $\color{#d91a1a}-1.07\%$
test_step_mdp_speed[True-False-True-True-False] 51.1960μs 21.4536μs 46.6123 KOps/s 46.5342 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-False-True-False-True] 72.8970μs 18.5882μs 53.7976 KOps/s 53.2501 KOps/s $\color{#35bf28}+1.03\%$
test_step_mdp_speed[True-False-True-False-False] 54.8620μs 11.8717μs 84.2342 KOps/s 84.5431 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-False-False-True-True] 67.7370μs 35.4401μs 28.2166 KOps/s 28.1251 KOps/s $\color{#35bf28}+0.33\%$
test_step_mdp_speed[True-False-False-True-False] 71.9750μs 23.2599μs 42.9925 KOps/s 43.2648 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[True-False-False-False-True] 49.0930μs 20.4581μs 48.8803 KOps/s 48.2632 KOps/s $\color{#35bf28}+1.28\%$
test_step_mdp_speed[True-False-False-False-False] 71.7040μs 13.5248μs 73.9382 KOps/s 73.4300 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[False-True-True-True-True] 85.4210μs 33.9218μs 29.4796 KOps/s 29.4392 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[False-True-True-True-False] 47.8800μs 21.5626μs 46.3766 KOps/s 45.6469 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[False-True-True-False-True] 72.9470μs 21.4019μs 46.7247 KOps/s 46.3023 KOps/s $\color{#35bf28}+0.91\%$
test_step_mdp_speed[False-True-True-False-False] 0.5786ms 13.3570μs 74.8674 KOps/s 75.4072 KOps/s $\color{#d91a1a}-0.72\%$
test_step_mdp_speed[False-True-False-True-True] 88.6890μs 35.2126μs 28.3989 KOps/s 28.1302 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[False-True-False-True-False] 73.1740μs 23.0829μs 43.3220 KOps/s 43.1740 KOps/s $\color{#35bf28}+0.34\%$
test_step_mdp_speed[False-True-False-False-True] 2.7289ms 23.2748μs 42.9650 KOps/s 43.1010 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[False-True-False-False-False] 65.8230μs 15.0291μs 66.5376 KOps/s 66.5634 KOps/s $\color{#d91a1a}-0.04\%$
test_step_mdp_speed[False-False-True-True-True] 87.5440μs 37.3083μs 26.8037 KOps/s 26.6305 KOps/s $\color{#35bf28}+0.65\%$
test_step_mdp_speed[False-False-True-True-False] 60.3940μs 25.0205μs 39.9672 KOps/s 39.9484 KOps/s $\color{#35bf28}+0.05\%$
test_step_mdp_speed[False-False-True-False-True] 60.1120μs 23.1177μs 43.2568 KOps/s 44.0097 KOps/s $\color{#d91a1a}-1.71\%$
test_step_mdp_speed[False-False-True-False-False] 41.4880μs 14.9185μs 67.0309 KOps/s 67.2361 KOps/s $\color{#d91a1a}-0.31\%$
test_step_mdp_speed[False-False-False-True-True] 94.6780μs 38.7611μs 25.7991 KOps/s 25.8937 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[False-False-False-True-False] 64.8120μs 26.7059μs 37.4449 KOps/s 37.2376 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-False-False-False-True] 75.7020μs 24.8053μs 40.3140 KOps/s 40.8446 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[False-False-False-False-False] 73.4180μs 16.5869μs 60.2886 KOps/s 60.4954 KOps/s $\color{#d91a1a}-0.34\%$
test_values[generalized_advantage_estimate-True-True] 12.0020ms 9.7775ms 102.2760 Ops/s 103.0998 Ops/s $\color{#d91a1a}-0.80\%$
test_values[vec_generalized_advantage_estimate-True-True] 26.3016ms 24.2398ms 41.2545 Ops/s 37.9941 Ops/s $\textbf{\color{#35bf28}+8.58\%}$
test_values[td0_return_estimate-False-False] 0.2775ms 0.1839ms 5.4372 KOps/s 5.5992 KOps/s $\color{#d91a1a}-2.89\%$
test_values[td1_return_estimate-False-False] 26.1923ms 24.4352ms 40.9246 Ops/s 41.1882 Ops/s $\color{#d91a1a}-0.64\%$
test_values[vec_td1_return_estimate-False-False] 25.9672ms 24.2558ms 41.2272 Ops/s 37.8659 Ops/s $\textbf{\color{#35bf28}+8.88\%}$
test_values[td_lambda_return_estimate-True-False] 37.9518ms 35.2026ms 28.4070 Ops/s 28.2170 Ops/s $\color{#35bf28}+0.67\%$
test_values[vec_td_lambda_return_estimate-True-False] 26.2546ms 24.4050ms 40.9752 Ops/s 38.0483 Ops/s $\textbf{\color{#35bf28}+7.69\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.6714ms 8.5640ms 116.7683 Ops/s 117.2812 Ops/s $\color{#d91a1a}-0.44\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3296ms 1.9059ms 524.6928 Ops/s 513.8270 Ops/s $\color{#35bf28}+2.11\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6468ms 0.3722ms 2.6866 KOps/s 2.6797 KOps/s $\color{#35bf28}+0.26\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 40.2676ms 37.4997ms 26.6669 Ops/s 24.7003 Ops/s $\textbf{\color{#35bf28}+7.96\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.4637ms 3.4344ms 291.1745 Ops/s 274.6373 Ops/s $\textbf{\color{#35bf28}+6.02\%}$
test_dqn_speed[False-None] 6.3731ms 1.4163ms 706.0635 Ops/s 686.8340 Ops/s $\color{#35bf28}+2.80\%$
test_dqn_speed[False-backward] 1.9669ms 1.9054ms 524.8285 Ops/s 516.3188 Ops/s $\color{#35bf28}+1.65\%$
test_dqn_speed[True-None] 0.7766ms 0.5085ms 1.9666 KOps/s 1.9550 KOps/s $\color{#35bf28}+0.59\%$
test_dqn_speed[True-backward] 0.9989ms 0.9324ms 1.0725 KOps/s 1.0660 KOps/s $\color{#35bf28}+0.62\%$
test_dqn_speed[reduce-overhead-None] 0.7679ms 0.5057ms 1.9776 KOps/s 1.9973 KOps/s $\color{#d91a1a}-0.98\%$
test_dqn_speed[reduce-overhead-backward] 0.9910ms 0.9253ms 1.0807 KOps/s 1.0783 KOps/s $\color{#35bf28}+0.22\%$
test_ddpg_speed[False-None] 3.2703ms 2.9338ms 340.8585 Ops/s 339.9828 Ops/s $\color{#35bf28}+0.26\%$
test_ddpg_speed[False-backward] 4.2169ms 4.0767ms 245.2978 Ops/s 245.3499 Ops/s $\color{#d91a1a}-0.02\%$
test_ddpg_speed[True-None] 2.5501ms 1.2578ms 795.0123 Ops/s 790.6528 Ops/s $\color{#35bf28}+0.55\%$
test_ddpg_speed[True-backward] 2.2170ms 2.1471ms 465.7540 Ops/s 464.2049 Ops/s $\color{#35bf28}+0.33\%$
test_ddpg_speed[reduce-overhead-None] 1.5331ms 1.2531ms 798.0509 Ops/s 784.6786 Ops/s $\color{#35bf28}+1.70\%$
test_ddpg_speed[reduce-overhead-backward] 2.1936ms 2.1432ms 466.5990 Ops/s 463.8383 Ops/s $\color{#35bf28}+0.60\%$
test_sac_speed[False-None] 9.2416ms 8.1932ms 122.0529 Ops/s 122.8262 Ops/s $\color{#d91a1a}-0.63\%$
test_sac_speed[False-backward] 12.7394ms 11.0302ms 90.6600 Ops/s 91.7652 Ops/s $\color{#d91a1a}-1.20\%$
test_sac_speed[True-None] 2.6926ms 2.1773ms 459.2855 Ops/s 463.3036 Ops/s $\color{#d91a1a}-0.87\%$
test_sac_speed[True-backward] 3.9448ms 3.8172ms 261.9753 Ops/s 256.3157 Ops/s $\color{#35bf28}+2.21\%$
test_sac_speed[reduce-overhead-None] 2.7262ms 2.1436ms 466.4973 Ops/s 463.8504 Ops/s $\color{#35bf28}+0.57\%$
test_sac_speed[reduce-overhead-backward] 3.9026ms 3.8137ms 262.2119 Ops/s 260.2801 Ops/s $\color{#35bf28}+0.74\%$
test_redq_speed[False-None] 14.1652ms 13.2412ms 75.5219 Ops/s 52.4788 Ops/s $\textbf{\color{#35bf28}+43.91\%}$
test_redq_speed[False-backward] 23.8726ms 22.6616ms 44.1275 Ops/s 39.7891 Ops/s $\textbf{\color{#35bf28}+10.90\%}$
test_redq_speed[True-None] 5.9772ms 5.1110ms 195.6553 Ops/s 193.3910 Ops/s $\color{#35bf28}+1.17\%$
test_redq_speed[True-backward] 13.0418ms 12.5862ms 79.4519 Ops/s 77.9943 Ops/s $\color{#35bf28}+1.87\%$
test_redq_speed[reduce-overhead-None] 5.7908ms 4.9637ms 201.4628 Ops/s 184.8930 Ops/s $\textbf{\color{#35bf28}+8.96\%}$
test_redq_speed[reduce-overhead-backward] 13.3192ms 12.5678ms 79.5685 Ops/s 78.3534 Ops/s $\color{#35bf28}+1.55\%$
test_redq_deprec_speed[False-None] 15.0478ms 12.9414ms 77.2714 Ops/s 75.1222 Ops/s $\color{#35bf28}+2.86\%$
test_redq_deprec_speed[False-backward] 19.9894ms 18.6826ms 53.5259 Ops/s 51.9384 Ops/s $\color{#35bf28}+3.06\%$
test_redq_deprec_speed[True-None] 5.1561ms 3.8772ms 257.9206 Ops/s 250.5961 Ops/s $\color{#35bf28}+2.92\%$
test_redq_deprec_speed[True-backward] 9.1152ms 8.5311ms 117.2186 Ops/s 115.1315 Ops/s $\color{#35bf28}+1.81\%$
test_redq_deprec_speed[reduce-overhead-None] 4.4131ms 3.9000ms 256.4131 Ops/s 252.3384 Ops/s $\color{#35bf28}+1.61\%$
test_redq_deprec_speed[reduce-overhead-backward] 9.3422ms 8.5969ms 116.3215 Ops/s 116.8481 Ops/s $\color{#d91a1a}-0.45\%$
test_td3_speed[False-None] 8.3347ms 8.0547ms 124.1518 Ops/s 122.6670 Ops/s $\color{#35bf28}+1.21\%$
test_td3_speed[False-backward] 10.9515ms 10.4310ms 95.8677 Ops/s 93.6809 Ops/s $\color{#35bf28}+2.33\%$
test_td3_speed[True-None] 1.9872ms 1.8150ms 550.9543 Ops/s 526.3747 Ops/s $\color{#35bf28}+4.67\%$
test_td3_speed[True-backward] 3.9560ms 3.4382ms 290.8497 Ops/s 278.1615 Ops/s $\color{#35bf28}+4.56\%$
test_td3_speed[reduce-overhead-None] 2.0229ms 1.8048ms 554.0668 Ops/s 528.2933 Ops/s $\color{#35bf28}+4.88\%$
test_td3_speed[reduce-overhead-backward] 3.4748ms 3.3882ms 295.1456 Ops/s 289.8711 Ops/s $\color{#35bf28}+1.82\%$
test_cql_speed[False-None] 37.5754ms 36.0291ms 27.7554 Ops/s 27.0727 Ops/s $\color{#35bf28}+2.52\%$
test_cql_speed[False-backward] 47.7445ms 46.0772ms 21.7027 Ops/s 21.0215 Ops/s $\color{#35bf28}+3.24\%$
test_cql_speed[True-None] 16.5971ms 15.9863ms 62.5537 Ops/s 61.2010 Ops/s $\color{#35bf28}+2.21\%$
test_cql_speed[True-backward] 24.4417ms 22.9864ms 43.5040 Ops/s 43.3441 Ops/s $\color{#35bf28}+0.37\%$
test_cql_speed[reduce-overhead-None] 17.5063ms 16.1516ms 61.9134 Ops/s 61.4798 Ops/s $\color{#35bf28}+0.71\%$
test_cql_speed[reduce-overhead-backward] 23.6820ms 22.7378ms 43.9796 Ops/s 42.4190 Ops/s $\color{#35bf28}+3.68\%$
test_a2c_speed[False-None] 8.0638ms 7.1262ms 140.3269 Ops/s 135.1025 Ops/s $\color{#35bf28}+3.87\%$
test_a2c_speed[False-backward] 15.8224ms 14.1891ms 70.4767 Ops/s 67.0550 Ops/s $\textbf{\color{#35bf28}+5.10\%}$
test_a2c_speed[True-None] 4.7132ms 3.7378ms 267.5361 Ops/s 264.2419 Ops/s $\color{#35bf28}+1.25\%$
test_a2c_speed[True-backward] 10.8656ms 10.2704ms 97.3668 Ops/s 96.6437 Ops/s $\color{#35bf28}+0.75\%$
test_a2c_speed[reduce-overhead-None] 4.6086ms 3.7118ms 269.4101 Ops/s 265.2069 Ops/s $\color{#35bf28}+1.58\%$
test_a2c_speed[reduce-overhead-backward] 11.6314ms 10.3032ms 97.0569 Ops/s 93.5318 Ops/s $\color{#35bf28}+3.77\%$
test_ppo_speed[False-None] 8.5415ms 7.4559ms 134.1220 Ops/s 130.1488 Ops/s $\color{#35bf28}+3.05\%$
test_ppo_speed[False-backward] 16.1855ms 15.0819ms 66.3048 Ops/s 65.6841 Ops/s $\color{#35bf28}+0.94\%$
test_ppo_speed[True-None] 4.6453ms 4.1119ms 243.1975 Ops/s 238.7632 Ops/s $\color{#35bf28}+1.86\%$
test_ppo_speed[True-backward] 12.4197ms 10.1959ms 98.0790 Ops/s 98.4217 Ops/s $\color{#d91a1a}-0.35\%$
test_ppo_speed[reduce-overhead-None] 4.7020ms 4.1027ms 243.7419 Ops/s 240.9593 Ops/s $\color{#35bf28}+1.15\%$
test_ppo_speed[reduce-overhead-backward] 11.2594ms 10.2000ms 98.0389 Ops/s 98.0013 Ops/s $\color{#35bf28}+0.04\%$
test_reinforce_speed[False-None] 8.1607ms 6.5516ms 152.6345 Ops/s 150.3408 Ops/s $\color{#35bf28}+1.53\%$
test_reinforce_speed[False-backward] 10.4367ms 9.8841ms 101.1730 Ops/s 100.8898 Ops/s $\color{#35bf28}+0.28\%$
test_reinforce_speed[True-None] 3.4752ms 3.0764ms 325.0593 Ops/s 317.7960 Ops/s $\color{#35bf28}+2.29\%$
test_reinforce_speed[True-backward] 9.9995ms 9.1423ms 109.3817 Ops/s 109.4439 Ops/s $\color{#d91a1a}-0.06\%$
test_reinforce_speed[reduce-overhead-None] 3.4011ms 3.0769ms 324.9999 Ops/s 317.1595 Ops/s $\color{#35bf28}+2.47\%$
test_reinforce_speed[reduce-overhead-backward] 9.7449ms 9.0812ms 110.1175 Ops/s 107.6198 Ops/s $\color{#35bf28}+2.32\%$
test_iql_speed[False-None] 33.4140ms 32.2560ms 31.0020 Ops/s 30.0658 Ops/s $\color{#35bf28}+3.11\%$
test_iql_speed[False-backward] 46.7379ms 45.2524ms 22.0983 Ops/s 21.5288 Ops/s $\color{#35bf28}+2.65\%$
test_iql_speed[True-None] 12.3658ms 11.3370ms 88.2069 Ops/s 85.9224 Ops/s $\color{#35bf28}+2.66\%$
test_iql_speed[True-backward] 27.5870ms 22.5185ms 44.4079 Ops/s 43.6419 Ops/s $\color{#35bf28}+1.76\%$
test_iql_speed[reduce-overhead-None] 12.2262ms 11.3199ms 88.3400 Ops/s 86.8142 Ops/s $\color{#35bf28}+1.76\%$
test_iql_speed[reduce-overhead-backward] 23.6708ms 22.6411ms 44.1675 Ops/s 43.5695 Ops/s $\color{#35bf28}+1.37\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9840ms 4.8514ms 206.1250 Ops/s 202.5776 Ops/s $\color{#35bf28}+1.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7155ms 0.5080ms 1.9687 KOps/s 1.9396 KOps/s $\color{#35bf28}+1.50\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7368ms 0.4894ms 2.0435 KOps/s 2.0165 KOps/s $\color{#35bf28}+1.34\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.5773ms 4.6731ms 213.9905 Ops/s 208.6128 Ops/s $\color{#35bf28}+2.58\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 3.2517ms 0.5054ms 1.9786 KOps/s 1.9672 KOps/s $\color{#35bf28}+0.58\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7078ms 0.4807ms 2.0801 KOps/s 2.0442 KOps/s $\color{#35bf28}+1.76\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.8991ms 1.6448ms 607.9919 Ops/s 598.8129 Ops/s $\color{#35bf28}+1.53\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.1114ms 1.5658ms 638.6418 Ops/s 634.5085 Ops/s $\color{#35bf28}+0.65\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.0674ms 4.8214ms 207.4078 Ops/s 199.4990 Ops/s $\color{#35bf28}+3.96\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.2333ms 0.6503ms 1.5377 KOps/s 1.5141 KOps/s $\color{#35bf28}+1.56\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9194ms 0.6286ms 1.5907 KOps/s 1.5939 KOps/s $\color{#d91a1a}-0.20\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.2195ms 4.7053ms 212.5280 Ops/s 210.3613 Ops/s $\color{#35bf28}+1.03\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2584ms 0.5099ms 1.9612 KOps/s 1.9104 KOps/s $\color{#35bf28}+2.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7410ms 0.4924ms 2.0309 KOps/s 2.0402 KOps/s $\color{#d91a1a}-0.45\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.4289ms 4.6403ms 215.5015 Ops/s 210.2356 Ops/s $\color{#35bf28}+2.50\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2732ms 0.5067ms 1.9737 KOps/s 1.9605 KOps/s $\color{#35bf28}+0.67\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6796ms 0.4810ms 2.0788 KOps/s 2.0280 KOps/s $\color{#35bf28}+2.51\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.0901ms 4.8002ms 208.3264 Ops/s 205.4432 Ops/s $\color{#35bf28}+1.40\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.7846ms 0.6524ms 1.5327 KOps/s 1.5338 KOps/s $\color{#d91a1a}-0.07\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8768ms 0.6270ms 1.5949 KOps/s 1.5707 KOps/s $\color{#35bf28}+1.54\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.6717ms 4.2050ms 237.8110 Ops/s 237.2777 Ops/s $\color{#35bf28}+0.22\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 6.4881ms 2.2985ms 435.0711 Ops/s 448.5531 Ops/s $\color{#d91a1a}-3.01\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.9056ms 1.4123ms 708.0758 Ops/s 728.0465 Ops/s $\color{#d91a1a}-2.74\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4272s 12.7378ms 78.5064 Ops/s 241.3259 Ops/s $\textbf{\color{#d91a1a}-67.47\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 6.6268ms 2.2836ms 437.9042 Ops/s 438.5116 Ops/s $\color{#d91a1a}-0.14\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.7674ms 1.2469ms 801.9940 Ops/s 633.5882 Ops/s $\textbf{\color{#35bf28}+26.58\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 5.6391ms 4.3364ms 230.6075 Ops/s 32.8745 Ops/s $\textbf{\color{#35bf28}+601.48\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.0368ms 2.4901ms 401.5856 Ops/s 392.8140 Ops/s $\color{#35bf28}+2.23\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.8421ms 1.5285ms 654.2471 Ops/s 613.4852 Ops/s $\textbf{\color{#35bf28}+6.64\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.5914ms 11.8665ms 84.2707 Ops/s 82.3171 Ops/s $\color{#35bf28}+2.37\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 15.8815ms 14.7373ms 67.8551 Ops/s 68.5198 Ops/s $\color{#d91a1a}-0.97\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 22.6930ms 20.7282ms 48.2434 Ops/s 48.0163 Ops/s $\color{#35bf28}+0.47\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 15.8761ms 14.8552ms 67.3165 Ops/s 66.9519 Ops/s $\color{#35bf28}+0.54\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 21.0622ms 20.6448ms 48.4384 Ops/s 47.8630 Ops/s $\color{#35bf28}+1.20\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 17.8593ms 16.1048ms 62.0933 Ops/s 61.8719 Ops/s $\color{#35bf28}+0.36\%$

Copy link

github-actions bot commented Feb 12, 2025

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

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

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.8756s 0.7890s 1.2675 Ops/s 1.2217 Ops/s $\color{#35bf28}+3.75\%$
test_transformed 1.4380s 1.3560s 0.7375 Ops/s 0.6975 Ops/s $\textbf{\color{#35bf28}+5.72\%}$
test_serial 2.3341s 2.2624s 0.4420 Ops/s 0.4280 Ops/s $\color{#35bf28}+3.28\%$
test_parallel 1.9391s 1.8635s 0.5366 Ops/s 0.5363 Ops/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[True-True-True-True-True] 0.1817ms 41.4935μs 24.1002 KOps/s 24.8995 KOps/s $\color{#d91a1a}-3.21\%$
test_step_mdp_speed[True-True-True-True-False] 93.0420μs 23.7935μs 42.0283 KOps/s 41.0726 KOps/s $\color{#35bf28}+2.33\%$
test_step_mdp_speed[True-True-True-False-True] 50.6410μs 23.1815μs 43.1378 KOps/s 44.3851 KOps/s $\color{#d91a1a}-2.81\%$
test_step_mdp_speed[True-True-True-False-False] 44.7010μs 13.2803μs 75.2997 KOps/s 74.6825 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[True-True-False-True-True] 70.0620μs 43.3879μs 23.0479 KOps/s 22.7834 KOps/s $\color{#35bf28}+1.16\%$
test_step_mdp_speed[True-True-False-True-False] 54.6210μs 26.1326μs 38.2664 KOps/s 37.9230 KOps/s $\color{#35bf28}+0.91\%$
test_step_mdp_speed[True-True-False-False-True] 0.1703ms 25.8652μs 38.6620 KOps/s 38.7971 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-True-False-False-False] 48.1210μs 15.7105μs 63.6516 KOps/s 63.0372 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-False-True-True-True] 78.5320μs 45.9661μs 21.7552 KOps/s 21.5001 KOps/s $\color{#35bf28}+1.19\%$
test_step_mdp_speed[True-False-True-True-False] 58.7210μs 28.0399μs 35.6635 KOps/s 34.5358 KOps/s $\color{#35bf28}+3.27\%$
test_step_mdp_speed[True-False-True-False-True] 55.0010μs 25.5308μs 39.1684 KOps/s 39.8776 KOps/s $\color{#d91a1a}-1.78\%$
test_step_mdp_speed[True-False-True-False-False] 49.2920μs 15.6321μs 63.9709 KOps/s 64.1565 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[True-False-False-True-True] 0.1147ms 47.5033μs 21.0512 KOps/s 20.5016 KOps/s $\color{#35bf28}+2.68\%$
test_step_mdp_speed[True-False-False-True-False] 55.0320μs 30.5884μs 32.6921 KOps/s 31.9708 KOps/s $\color{#35bf28}+2.26\%$
test_step_mdp_speed[True-False-False-False-True] 74.9420μs 27.7826μs 35.9938 KOps/s 36.4251 KOps/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-False-False-False-False] 46.3710μs 17.8603μs 55.9901 KOps/s 55.3702 KOps/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[False-True-True-True-True] 0.1069ms 45.8005μs 21.8338 KOps/s 22.1083 KOps/s $\color{#d91a1a}-1.24\%$
test_step_mdp_speed[False-True-True-True-False] 62.1120μs 28.3287μs 35.2998 KOps/s 34.6306 KOps/s $\color{#35bf28}+1.93\%$
test_step_mdp_speed[False-True-True-False-True] 92.0420μs 29.3659μs 34.0531 KOps/s 34.2373 KOps/s $\color{#d91a1a}-0.54\%$
test_step_mdp_speed[False-True-True-False-False] 0.2077ms 17.5677μs 56.9228 KOps/s 57.1619 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[False-True-False-True-True] 0.1449ms 48.5498μs 20.5974 KOps/s 20.7836 KOps/s $\color{#d91a1a}-0.90\%$
test_step_mdp_speed[False-True-False-True-False] 65.7510μs 30.6787μs 32.5959 KOps/s 32.4824 KOps/s $\color{#35bf28}+0.35\%$
test_step_mdp_speed[False-True-False-False-True] 3.3696ms 32.2211μs 31.0355 KOps/s 30.7907 KOps/s $\color{#35bf28}+0.80\%$
test_step_mdp_speed[False-True-False-False-False] 67.3410μs 19.7980μs 50.5102 KOps/s 49.6107 KOps/s $\color{#35bf28}+1.81\%$
test_step_mdp_speed[False-False-True-True-True] 0.1436ms 50.6962μs 19.7253 KOps/s 19.7785 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[False-False-True-True-False] 71.3110μs 33.0240μs 30.2810 KOps/s 29.6987 KOps/s $\color{#35bf28}+1.96\%$
test_step_mdp_speed[False-False-True-False-True] 69.2110μs 32.0716μs 31.1802 KOps/s 31.9050 KOps/s $\color{#d91a1a}-2.27\%$
test_step_mdp_speed[False-False-True-False-False] 0.1001ms 19.6698μs 50.8394 KOps/s 50.4657 KOps/s $\color{#35bf28}+0.74\%$
test_step_mdp_speed[False-False-False-True-True] 91.9320μs 52.1592μs 19.1721 KOps/s 18.8514 KOps/s $\color{#35bf28}+1.70\%$
test_step_mdp_speed[False-False-False-True-False] 71.8020μs 35.0849μs 28.5023 KOps/s 27.8084 KOps/s $\color{#35bf28}+2.50\%$
test_step_mdp_speed[False-False-False-False-True] 0.1729ms 33.7175μs 29.6582 KOps/s 29.7420 KOps/s $\color{#d91a1a}-0.28\%$
test_step_mdp_speed[False-False-False-False-False] 59.8010μs 21.4700μs 46.5766 KOps/s 45.1076 KOps/s $\color{#35bf28}+3.26\%$
test_values[generalized_advantage_estimate-True-True] 26.6806ms 25.2663ms 39.5784 Ops/s 39.5829 Ops/s $\color{#d91a1a}-0.01\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1036s 2.9567ms 338.2125 Ops/s 303.4859 Ops/s $\textbf{\color{#35bf28}+11.44\%}$
test_values[td0_return_estimate-False-False] 0.1031ms 78.7876μs 12.6924 KOps/s 12.0398 KOps/s $\textbf{\color{#35bf28}+5.42\%}$
test_values[td1_return_estimate-False-False] 59.5736ms 57.3890ms 17.4249 Ops/s 17.8999 Ops/s $\color{#d91a1a}-2.65\%$
test_values[vec_td1_return_estimate-False-False] 1.3042ms 1.0745ms 930.6718 Ops/s 932.0905 Ops/s $\color{#d91a1a}-0.15\%$
test_values[td_lambda_return_estimate-True-False] 95.0218ms 88.8044ms 11.2607 Ops/s 11.2615 Ops/s $-0.01\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2985ms 1.0670ms 937.2018 Ops/s 938.0136 Ops/s $\color{#d91a1a}-0.09\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.9891ms 26.5543ms 37.6586 Ops/s 40.1968 Ops/s $\textbf{\color{#d91a1a}-6.31\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0282ms 0.7588ms 1.3178 KOps/s 1.3381 KOps/s $\color{#d91a1a}-1.51\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7994ms 0.6521ms 1.5335 KOps/s 1.5119 KOps/s $\color{#35bf28}+1.43\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6351ms 1.4741ms 678.3825 Ops/s 674.9292 Ops/s $\color{#35bf28}+0.51\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8313ms 0.6656ms 1.5023 KOps/s 1.4797 KOps/s $\color{#35bf28}+1.53\%$
test_dqn_speed[False-None] 7.0668ms 1.5107ms 661.9380 Ops/s 643.0783 Ops/s $\color{#35bf28}+2.93\%$
test_dqn_speed[False-backward] 2.3211ms 2.1185ms 472.0265 Ops/s 463.6038 Ops/s $\color{#35bf28}+1.82\%$
test_dqn_speed[True-None] 0.7292ms 0.5649ms 1.7701 KOps/s 1.7057 KOps/s $\color{#35bf28}+3.78\%$
test_dqn_speed[True-backward] 1.4331ms 1.2469ms 801.9731 Ops/s 792.0421 Ops/s $\color{#35bf28}+1.25\%$
test_dqn_speed[reduce-overhead-None] 0.8037ms 0.5937ms 1.6844 KOps/s 1.6872 KOps/s $\color{#d91a1a}-0.16\%$
test_dqn_speed[reduce-overhead-backward] 1.2344ms 1.0857ms 921.0511 Ops/s 910.6160 Ops/s $\color{#35bf28}+1.15\%$
test_ddpg_speed[False-None] 3.1401ms 2.8571ms 350.0004 Ops/s 342.9483 Ops/s $\color{#35bf28}+2.06\%$
test_ddpg_speed[False-backward] 4.3551ms 4.1831ms 239.0585 Ops/s 233.7995 Ops/s $\color{#35bf28}+2.25\%$
test_ddpg_speed[True-None] 1.5522ms 1.3610ms 734.7739 Ops/s 729.5129 Ops/s $\color{#35bf28}+0.72\%$
test_ddpg_speed[True-backward] 2.5883ms 2.4455ms 408.9123 Ops/s 378.6987 Ops/s $\textbf{\color{#35bf28}+7.98\%}$
test_ddpg_speed[reduce-overhead-None] 1.5330ms 1.3680ms 731.0197 Ops/s 722.2499 Ops/s $\color{#35bf28}+1.21\%$
test_ddpg_speed[reduce-overhead-backward] 2.0818ms 1.9131ms 522.7072 Ops/s 480.2378 Ops/s $\textbf{\color{#35bf28}+8.84\%}$
test_sac_speed[False-None] 8.3515ms 7.9582ms 125.6563 Ops/s 121.9236 Ops/s $\color{#35bf28}+3.06\%$
test_sac_speed[False-backward] 11.3449ms 10.8305ms 92.3318 Ops/s 87.6109 Ops/s $\textbf{\color{#35bf28}+5.39\%}$
test_sac_speed[True-None] 2.1485ms 1.9538ms 511.8194 Ops/s 508.1807 Ops/s $\color{#35bf28}+0.72\%$
test_sac_speed[True-backward] 3.9148ms 3.6408ms 274.6679 Ops/s 266.0556 Ops/s $\color{#35bf28}+3.24\%$
test_sac_speed[reduce-overhead-None] 18.1763ms 10.8935ms 91.7980 Ops/s 91.3157 Ops/s $\color{#35bf28}+0.53\%$
test_sac_speed[reduce-overhead-backward] 1.8339ms 1.6592ms 602.6945 Ops/s 584.5546 Ops/s $\color{#35bf28}+3.10\%$
test_redq_speed[False-None] 8.1997ms 7.7017ms 129.8413 Ops/s 129.8643 Ops/s $\color{#d91a1a}-0.02\%$
test_redq_speed[False-backward] 12.1754ms 11.4509ms 87.3291 Ops/s 86.3298 Ops/s $\color{#35bf28}+1.16\%$
test_redq_speed[True-None] 2.5781ms 2.3473ms 426.0279 Ops/s 416.8032 Ops/s $\color{#35bf28}+2.21\%$
test_redq_speed[True-backward] 4.4022ms 4.0515ms 246.8207 Ops/s 229.0590 Ops/s $\textbf{\color{#35bf28}+7.75\%}$
test_redq_speed[reduce-overhead-None] 2.7934ms 2.4175ms 413.6466 Ops/s 413.4224 Ops/s $\color{#35bf28}+0.05\%$
test_redq_speed[reduce-overhead-backward] 4.3553ms 4.0707ms 245.6600 Ops/s 241.8860 Ops/s $\color{#35bf28}+1.56\%$
test_redq_deprec_speed[False-None] 9.5134ms 8.9618ms 111.5842 Ops/s 108.8089 Ops/s $\color{#35bf28}+2.55\%$
test_redq_deprec_speed[False-backward] 12.5225ms 11.9233ms 83.8692 Ops/s 82.1687 Ops/s $\color{#35bf28}+2.07\%$
test_redq_deprec_speed[True-None] 2.9331ms 2.7071ms 369.3955 Ops/s 365.1582 Ops/s $\color{#35bf28}+1.16\%$
test_redq_deprec_speed[True-backward] 4.7556ms 4.3695ms 228.8603 Ops/s 222.4649 Ops/s $\color{#35bf28}+2.87\%$
test_redq_deprec_speed[reduce-overhead-None] 2.9799ms 2.7102ms 368.9781 Ops/s 363.3368 Ops/s $\color{#35bf28}+1.55\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.4979ms 4.3282ms 231.0455 Ops/s 218.4970 Ops/s $\textbf{\color{#35bf28}+5.74\%}$
test_td3_speed[False-None] 8.1377ms 7.9945ms 125.0859 Ops/s 123.8892 Ops/s $\color{#35bf28}+0.97\%$
test_td3_speed[False-backward] 11.2592ms 10.2986ms 97.1009 Ops/s 93.6453 Ops/s $\color{#35bf28}+3.69\%$
test_td3_speed[True-None] 1.8241ms 1.7149ms 583.1352 Ops/s 574.7650 Ops/s $\color{#35bf28}+1.46\%$
test_td3_speed[True-backward] 3.3322ms 3.1962ms 312.8749 Ops/s 287.5480 Ops/s $\textbf{\color{#35bf28}+8.81\%}$
test_td3_speed[reduce-overhead-None] 70.8808ms 27.1513ms 36.8306 Ops/s 37.2793 Ops/s $\color{#d91a1a}-1.20\%$
test_td3_speed[reduce-overhead-backward] 1.5627ms 1.4164ms 706.0276 Ops/s 640.8170 Ops/s $\textbf{\color{#35bf28}+10.18\%}$
test_cql_speed[False-None] 17.5505ms 16.7282ms 59.7793 Ops/s 59.1021 Ops/s $\color{#35bf28}+1.15\%$
test_cql_speed[False-backward] 22.9065ms 21.7071ms 46.0679 Ops/s 44.6369 Ops/s $\color{#35bf28}+3.21\%$
test_cql_speed[True-None] 3.4827ms 3.2862ms 304.3073 Ops/s 294.1649 Ops/s $\color{#35bf28}+3.45\%$
test_cql_speed[True-backward] 5.9521ms 5.5192ms 181.1865 Ops/s 171.5886 Ops/s $\textbf{\color{#35bf28}+5.59\%}$
test_cql_speed[reduce-overhead-None] 19.8578ms 13.0989ms 76.3426 Ops/s 75.6997 Ops/s $\color{#35bf28}+0.85\%$
test_cql_speed[reduce-overhead-backward] 2.0243ms 1.8549ms 539.1236 Ops/s 525.1183 Ops/s $\color{#35bf28}+2.67\%$
test_a2c_speed[False-None] 3.3481ms 3.1410ms 318.3670 Ops/s 304.8094 Ops/s $\color{#35bf28}+4.45\%$
test_a2c_speed[False-backward] 6.9963ms 6.0046ms 166.5400 Ops/s 161.1236 Ops/s $\color{#35bf28}+3.36\%$
test_a2c_speed[True-None] 1.6340ms 1.3892ms 719.8539 Ops/s 729.7984 Ops/s $\color{#d91a1a}-1.36\%$
test_a2c_speed[True-backward] 3.1055ms 2.9333ms 340.9122 Ops/s 319.0936 Ops/s $\textbf{\color{#35bf28}+6.84\%}$
test_a2c_speed[reduce-overhead-None] 14.5548ms 8.5385ms 117.1168 Ops/s 120.7715 Ops/s $\color{#d91a1a}-3.03\%$
test_a2c_speed[reduce-overhead-backward] 1.6279ms 1.4779ms 676.6530 Ops/s 608.2108 Ops/s $\textbf{\color{#35bf28}+11.25\%}$
test_ppo_speed[False-None] 3.8602ms 3.6438ms 274.4356 Ops/s 264.1927 Ops/s $\color{#35bf28}+3.88\%$
test_ppo_speed[False-backward] 7.1620ms 6.6853ms 149.5809 Ops/s 139.4239 Ops/s $\textbf{\color{#35bf28}+7.28\%}$
test_ppo_speed[True-None] 1.6012ms 1.4252ms 701.6440 Ops/s 687.7641 Ops/s $\color{#35bf28}+2.02\%$
test_ppo_speed[True-backward] 3.3569ms 3.2058ms 311.9304 Ops/s 299.9335 Ops/s $\color{#35bf28}+4.00\%$
test_ppo_speed[reduce-overhead-None] 1.1790ms 0.9859ms 1.0143 KOps/s 1.0201 KOps/s $\color{#d91a1a}-0.57\%$
test_ppo_speed[reduce-overhead-backward] 1.5656ms 1.4284ms 700.0841 Ops/s 612.3959 Ops/s $\textbf{\color{#35bf28}+14.32\%}$
test_reinforce_speed[False-None] 2.5624ms 2.2613ms 442.2261 Ops/s 431.3451 Ops/s $\color{#35bf28}+2.52\%$
test_reinforce_speed[False-backward] 3.4813ms 3.2512ms 307.5765 Ops/s 287.8859 Ops/s $\textbf{\color{#35bf28}+6.84\%}$
test_reinforce_speed[True-None] 1.4773ms 1.3057ms 765.8645 Ops/s 745.6994 Ops/s $\color{#35bf28}+2.70\%$
test_reinforce_speed[True-backward] 3.0972ms 2.9463ms 339.4084 Ops/s 321.8953 Ops/s $\textbf{\color{#35bf28}+5.44\%}$
test_reinforce_speed[reduce-overhead-None] 15.7751ms 8.9474ms 111.7644 Ops/s 108.8252 Ops/s $\color{#35bf28}+2.70\%$
test_reinforce_speed[reduce-overhead-backward] 1.6534ms 1.5188ms 658.4016 Ops/s 589.6144 Ops/s $\textbf{\color{#35bf28}+11.67\%}$
test_iql_speed[False-None] 9.8141ms 9.1119ms 109.7469 Ops/s 106.0811 Ops/s $\color{#35bf28}+3.46\%$
test_iql_speed[False-backward] 13.2874ms 12.7106ms 78.6743 Ops/s 74.3458 Ops/s $\textbf{\color{#35bf28}+5.82\%}$
test_iql_speed[True-None] 2.4210ms 2.2506ms 444.3174 Ops/s 417.7044 Ops/s $\textbf{\color{#35bf28}+6.37\%}$
test_iql_speed[True-backward] 4.9990ms 4.7901ms 208.7654 Ops/s 194.6780 Ops/s $\textbf{\color{#35bf28}+7.24\%}$
test_iql_speed[reduce-overhead-None] 0.4823s 12.7761ms 78.2711 Ops/s 94.8622 Ops/s $\textbf{\color{#d91a1a}-17.49\%}$
test_iql_speed[reduce-overhead-backward] 2.0675ms 1.9398ms 515.5139 Ops/s 458.5221 Ops/s $\textbf{\color{#35bf28}+12.43\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.8268ms 6.3574ms 157.2976 Ops/s 154.3596 Ops/s $\color{#35bf28}+1.90\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5173ms 0.2716ms 3.6825 KOps/s 3.6570 KOps/s $\color{#35bf28}+0.70\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7059ms 0.2518ms 3.9710 KOps/s 3.9594 KOps/s $\color{#35bf28}+0.29\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.5963ms 6.1453ms 162.7263 Ops/s 163.4756 Ops/s $\color{#d91a1a}-0.46\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8336ms 0.2921ms 3.4236 KOps/s 3.2378 KOps/s $\textbf{\color{#35bf28}+5.74\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5173ms 0.2727ms 3.6670 KOps/s 3.1826 KOps/s $\textbf{\color{#35bf28}+15.22\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6186ms 1.3155ms 760.1758 Ops/s 715.6010 Ops/s $\textbf{\color{#35bf28}+6.23\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4541ms 1.1971ms 835.3579 Ops/s 769.4013 Ops/s $\textbf{\color{#35bf28}+8.57\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6033ms 6.2702ms 159.4851 Ops/s 157.9959 Ops/s $\color{#35bf28}+0.94\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9543ms 0.4755ms 2.1031 KOps/s 2.2963 KOps/s $\textbf{\color{#d91a1a}-8.41\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8637ms 0.4780ms 2.0921 KOps/s 2.5187 KOps/s $\textbf{\color{#d91a1a}-16.94\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.4439ms 6.1476ms 162.6662 Ops/s 162.7623 Ops/s $\color{#d91a1a}-0.06\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.1013ms 0.3489ms 2.8658 KOps/s 3.6072 KOps/s $\textbf{\color{#d91a1a}-20.55\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6972ms 0.2512ms 3.9811 KOps/s 3.9677 KOps/s $\color{#35bf28}+0.34\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.4494ms 6.1093ms 163.6841 Ops/s 162.2184 Ops/s $\color{#35bf28}+0.90\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0532ms 0.3015ms 3.3166 KOps/s 2.9635 KOps/s $\textbf{\color{#35bf28}+11.92\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5804ms 0.3011ms 3.3213 KOps/s 3.0896 KOps/s $\textbf{\color{#35bf28}+7.50\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6112ms 6.3023ms 158.6713 Ops/s 157.0831 Ops/s $\color{#35bf28}+1.01\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1964ms 0.4222ms 2.3686 KOps/s 2.1697 KOps/s $\textbf{\color{#35bf28}+9.16\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7127ms 0.3998ms 2.5015 KOps/s 2.0328 KOps/s $\textbf{\color{#35bf28}+23.06\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0581ms 5.5285ms 180.8809 Ops/s 178.7076 Ops/s $\color{#35bf28}+1.22\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.7886ms 1.8435ms 542.4481 Ops/s 429.1078 Ops/s $\textbf{\color{#35bf28}+26.41\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.7142ms 1.2685ms 788.3394 Ops/s 828.9689 Ops/s $\color{#d91a1a}-4.90\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4665s 14.8602ms 67.2940 Ops/s 180.2224 Ops/s $\textbf{\color{#d91a1a}-62.66\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.3881ms 2.0452ms 488.9425 Ops/s 421.2456 Ops/s $\textbf{\color{#35bf28}+16.07\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1327ms 1.2726ms 785.8125 Ops/s 848.7215 Ops/s $\textbf{\color{#d91a1a}-7.41\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.3736ms 5.7569ms 173.7040 Ops/s 30.7592 Ops/s $\textbf{\color{#35bf28}+464.72\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.8426ms 2.1904ms 456.5479 Ops/s 455.6892 Ops/s $\color{#35bf28}+0.19\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.8672ms 1.3909ms 718.9756 Ops/s 721.3684 Ops/s $\color{#d91a1a}-0.33\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.1773ms 12.8928ms 77.5628 Ops/s 72.4626 Ops/s $\textbf{\color{#35bf28}+7.04\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.4284ms 16.6890ms 59.9195 Ops/s 59.4251 Ops/s $\color{#35bf28}+0.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 17.9075ms 17.7189ms 56.4368 Ops/s 53.5816 Ops/s $\textbf{\color{#35bf28}+5.33\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.5881ms 16.9299ms 59.0670 Ops/s 58.6512 Ops/s $\color{#35bf28}+0.71\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 0.3984s 25.4450ms 39.3005 Ops/s 54.3569 Ops/s $\textbf{\color{#d91a1a}-27.70\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.8315ms 18.4326ms 54.2517 Ops/s 54.2616 Ops/s $\color{#d91a1a}-0.02\%$

@vmoens vmoens added the bug Something isn't working label Feb 12, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 12, 2025
ghstack-source-id: 38b90fe50eb6ad6849c1d47e507098565c08f769
Pull Request resolved: #2784
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: 6003a85afde544cdac8f5bf65d865a17f2e53f69
Pull Request resolved: #2784
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens merged commit 46c7263 into gh/vmoens/89/base Feb 13, 2025
71 of 74 checks passed
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: 23fe46b61dc2c9548fd9de7e4100431918fd0370
Pull Request resolved: #2784
@vmoens vmoens deleted the gh/vmoens/89/head branch February 13, 2025 17:44
vmoens added a commit that referenced this pull request Feb 17, 2025
ghstack-source-id: 23fe46b61dc2c9548fd9de7e4100431918fd0370
Pull Request resolved: #2784

(cherry picked from commit e084c02)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working 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