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.

Sign up for GitHub

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

Jump to bottom

[BugFix] Avoid KeyError in slice sampler (for compile) #2670

Merged
merged 2 commits into from
Dec 20, 2024
Merged

[BugFix] Avoid KeyError in slice sampler (for compile) #2670

merged 2 commits into from
Dec 20, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Dec 19, 2024 •
edited
Loading

@vmoens
Update
6a7ad64 
[ghstack-poisoned]
@vmoens vmoens mentioned this pull request Dec 19, 2024
Merged
@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Dec 19, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2670

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

❌ 8 New Failures, 7 Unrelated Failures

As of commit 9a60ec9 with merge base 133d709 (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following jobs 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.

vmoens added a commit that referenced this pull request Dec 19, 2024
@vmoens
[BugFix] Avoid KeyError in slice sampler (for compile)
bf5ffff 
ghstack-source-id: 75d8882254aad3c6d8c2bb5c993a5fe93c9143e8
Pull Request resolved: #2670
@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 Dec 19, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Dec 19, 2024 •
edited
Loading

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4255s 0.4245s 2.3555 Ops/s 2.2113 Ops/s $\textbf{\color{#35bf28}+6.52\%}$
test_transformed 0.6027s 0.6005s 1.6653 Ops/s 1.6193 Ops/s $\color{#35bf28}+2.84\%$
test_serial 1.3439s 1.3415s 0.7454 Ops/s 0.7194 Ops/s $\color{#35bf28}+3.62\%$
test_parallel 1.2794s 1.2042s 0.8305 Ops/s 0.8130 Ops/s $\color{#35bf28}+2.15\%$
test_step_mdp_speed[True-True-True-True-True] 0.1531ms 31.2173μs 32.0336 KOps/s 31.8709 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-True-True-True-False] 49.0820μs 17.9567μs 55.6894 KOps/s 54.6995 KOps/s $\color{#35bf28}+1.81\%$
test_step_mdp_speed[True-True-True-False-True] 61.8250μs 17.6455μs 56.6716 KOps/s 55.0005 KOps/s $\color{#35bf28}+3.04\%$
test_step_mdp_speed[True-True-True-False-False] 36.7090μs 10.1676μs 98.3514 KOps/s 95.3851 KOps/s $\color{#35bf28}+3.11\%$
test_step_mdp_speed[True-True-False-True-True] 74.8400μs 33.4690μs 29.8784 KOps/s 29.5193 KOps/s $\color{#35bf28}+1.22\%$
test_step_mdp_speed[True-True-False-True-False] 64.0390μs 20.3211μs 49.2100 KOps/s 48.3915 KOps/s $\color{#35bf28}+1.69\%$
test_step_mdp_speed[True-True-False-False-True] 50.4640μs 19.7820μs 50.5509 KOps/s 49.0533 KOps/s $\color{#35bf28}+3.05\%$
test_step_mdp_speed[True-True-False-False-False] 49.7930μs 12.0990μs 82.6514 KOps/s 79.4925 KOps/s $\color{#35bf28}+3.97\%$
test_step_mdp_speed[True-False-True-True-True] 92.6230μs 35.4735μs 28.1900 KOps/s 27.8278 KOps/s $\color{#35bf28}+1.30\%$
test_step_mdp_speed[True-False-True-True-False] 74.8100μs 22.2205μs 45.0035 KOps/s 43.9427 KOps/s $\color{#35bf28}+2.41\%$
test_step_mdp_speed[True-False-True-False-True] 96.8920μs 19.1226μs 52.2942 KOps/s 49.8752 KOps/s $\color{#35bf28}+4.85\%$
test_step_mdp_speed[True-False-True-False-False] 44.4730μs 12.0372μs 83.0760 KOps/s 79.8466 KOps/s $\color{#35bf28}+4.04\%$
test_step_mdp_speed[True-False-False-True-True] 0.3458ms 38.3289μs 26.0899 KOps/s 26.4644 KOps/s $\color{#d91a1a}-1.41\%$
test_step_mdp_speed[True-False-False-True-False] 53.2090μs 23.8918μs 41.8553 KOps/s 40.9387 KOps/s $\color{#35bf28}+2.24\%$
test_step_mdp_speed[True-False-False-False-True] 50.0430μs 21.4221μs 46.6808 KOps/s 45.5243 KOps/s $\color{#35bf28}+2.54\%$
test_step_mdp_speed[True-False-False-False-False] 40.3760μs 13.8262μs 72.3263 KOps/s 69.7944 KOps/s $\color{#35bf28}+3.63\%$
test_step_mdp_speed[False-True-True-True-True] 77.4340μs 35.1684μs 28.4346 KOps/s 25.6754 KOps/s $\textbf{\color{#35bf28}+10.75\%}$
test_step_mdp_speed[False-True-True-True-False] 55.3230μs 22.0966μs 45.2558 KOps/s 44.5548 KOps/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[False-True-True-False-True] 0.1916ms 22.4533μs 44.5369 KOps/s 42.8917 KOps/s $\color{#35bf28}+3.84\%$
test_step_mdp_speed[False-True-True-False-False] 0.1386ms 13.6607μs 73.2026 KOps/s 72.5164 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[False-True-False-True-True] 86.9120μs 37.4879μs 26.6753 KOps/s 26.4686 KOps/s $\color{#35bf28}+0.78\%$
test_step_mdp_speed[False-True-False-True-False] 56.9360μs 23.9853μs 41.6922 KOps/s 41.2461 KOps/s $\color{#35bf28}+1.08\%$
test_step_mdp_speed[False-True-False-False-True] 2.7076ms 24.5806μs 40.6825 KOps/s 40.1849 KOps/s $\color{#35bf28}+1.24\%$
test_step_mdp_speed[False-True-False-False-False] 66.2910μs 15.3268μs 65.2452 KOps/s 63.6699 KOps/s $\color{#35bf28}+2.47\%$
test_step_mdp_speed[False-False-True-True-True] 0.1129ms 38.4794μs 25.9879 KOps/s 25.5076 KOps/s $\color{#35bf28}+1.88\%$
test_step_mdp_speed[False-False-True-True-False] 75.5900μs 25.5341μs 39.1634 KOps/s 38.2757 KOps/s $\color{#35bf28}+2.32\%$
test_step_mdp_speed[False-False-True-False-True] 58.0680μs 24.3540μs 41.0610 KOps/s 37.0103 KOps/s $\textbf{\color{#35bf28}+10.94\%}$
test_step_mdp_speed[False-False-True-False-False] 37.9400μs 15.5840μs 64.1686 KOps/s 64.4935 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[False-False-False-True-True] 0.1031ms 40.5928μs 24.6349 KOps/s 24.6403 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[False-False-False-True-False] 57.2970μs 27.4721μs 36.4006 KOps/s 35.9597 KOps/s $\color{#35bf28}+1.23\%$
test_step_mdp_speed[False-False-False-False-True] 70.6020μs 26.3841μs 37.9016 KOps/s 38.3484 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[False-False-False-False-False] 0.1465ms 17.4940μs 57.1624 KOps/s 57.5285 KOps/s $\color{#d91a1a}-0.64\%$
test_values[generalized_advantage_estimate-True-True] 15.3534ms 10.0078ms 99.9223 Ops/s 103.2455 Ops/s $\color{#d91a1a}-3.22\%$
test_values[vec_generalized_advantage_estimate-True-True] 36.7722ms 33.3816ms 29.9567 Ops/s 29.3812 Ops/s $\color{#35bf28}+1.96\%$
test_values[td0_return_estimate-False-False] 0.2533ms 0.1746ms 5.7282 KOps/s 5.7363 KOps/s $\color{#d91a1a}-0.14\%$
test_values[td1_return_estimate-False-False] 25.0171ms 24.4559ms 40.8900 Ops/s 41.2495 Ops/s $\color{#d91a1a}-0.87\%$
test_values[vec_td1_return_estimate-False-False] 37.2098ms 33.7624ms 29.6187 Ops/s 29.8676 Ops/s $\color{#d91a1a}-0.83\%$
test_values[td_lambda_return_estimate-True-False] 38.7812ms 35.4325ms 28.2227 Ops/s 28.8015 Ops/s $\color{#d91a1a}-2.01\%$
test_values[vec_td_lambda_return_estimate-True-False] 37.8418ms 33.9730ms 29.4351 Ops/s 30.0617 Ops/s $\color{#d91a1a}-2.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.4004ms 8.5024ms 117.6136 Ops/s 117.3645 Ops/s $\color{#35bf28}+0.21\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.2990ms 1.8793ms 532.1216 Ops/s 579.6890 Ops/s $\textbf{\color{#d91a1a}-8.21\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6217ms 0.3599ms 2.7786 KOps/s 2.8333 KOps/s $\color{#d91a1a}-1.93\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 42.4678ms 38.4869ms 25.9829 Ops/s 27.0781 Ops/s $\color{#d91a1a}-4.04\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.0561ms 3.0415ms 328.7865 Ops/s 331.5302 Ops/s $\color{#d91a1a}-0.83\%$
test_dqn_speed[False-None] 5.3743ms 1.4079ms 710.2776 Ops/s 717.2930 Ops/s $\color{#d91a1a}-0.98\%$
test_dqn_speed[False-backward] 3.0073ms 2.0520ms 487.3238 Ops/s 535.6567 Ops/s $\textbf{\color{#d91a1a}-9.02\%}$
test_dqn_speed[True-None] 0.6642ms 0.4845ms 2.0638 KOps/s 2.0458 KOps/s $\color{#35bf28}+0.88\%$
test_dqn_speed[True-backward] 1.0535ms 0.9191ms 1.0880 KOps/s 1.0975 KOps/s $\color{#d91a1a}-0.86\%$
test_dqn_speed[reduce-overhead-None] 0.6735ms 0.4933ms 2.0273 KOps/s 2.0747 KOps/s $\color{#d91a1a}-2.28\%$
test_dqn_speed[reduce-overhead-backward] 0.9575ms 0.9066ms 1.1030 KOps/s 1.0880 KOps/s $\color{#35bf28}+1.38\%$
test_ddpg_speed[False-None] 3.3252ms 2.9449ms 339.5752 Ops/s 344.6899 Ops/s $\color{#d91a1a}-1.48\%$
test_ddpg_speed[False-backward] 4.3914ms 4.0693ms 245.7425 Ops/s 248.6495 Ops/s $\color{#d91a1a}-1.17\%$
test_ddpg_speed[True-None] 1.3208ms 1.0174ms 982.9419 Ops/s 975.1394 Ops/s $\color{#35bf28}+0.80\%$
test_ddpg_speed[True-backward] 2.0048ms 1.9383ms 515.9050 Ops/s 525.8578 Ops/s $\color{#d91a1a}-1.89\%$
test_ddpg_speed[reduce-overhead-None] 1.3843ms 1.0186ms 981.7210 Ops/s 985.7480 Ops/s $\color{#d91a1a}-0.41\%$
test_ddpg_speed[reduce-overhead-backward] 1.9590ms 1.9045ms 525.0593 Ops/s 523.0736 Ops/s $\color{#35bf28}+0.38\%$
test_sac_speed[False-None] 8.6012ms 8.1042ms 123.3922 Ops/s 121.6608 Ops/s $\color{#35bf28}+1.42\%$
test_sac_speed[False-backward] 12.9367ms 10.9196ms 91.5787 Ops/s 93.0256 Ops/s $\color{#d91a1a}-1.56\%$
test_sac_speed[True-None] 2.1619ms 1.8430ms 542.6047 Ops/s 540.5322 Ops/s $\color{#35bf28}+0.38\%$
test_sac_speed[True-backward] 3.6212ms 3.5160ms 284.4135 Ops/s 275.2520 Ops/s $\color{#35bf28}+3.33\%$
test_sac_speed[reduce-overhead-None] 2.7762ms 1.8625ms 536.9269 Ops/s 529.6039 Ops/s $\color{#35bf28}+1.38\%$
test_sac_speed[reduce-overhead-backward] 3.6505ms 3.5279ms 283.4534 Ops/s 274.7366 Ops/s $\color{#35bf28}+3.17\%$
test_redq_speed[False-None] 15.0465ms 12.7757ms 78.2735 Ops/s 75.7750 Ops/s $\color{#35bf28}+3.30\%$
test_redq_speed[False-backward] 24.7580ms 22.2785ms 44.8863 Ops/s 44.5408 Ops/s $\color{#35bf28}+0.78\%$
test_redq_speed[True-None] 5.4297ms 4.5143ms 221.5194 Ops/s 199.5400 Ops/s $\textbf{\color{#35bf28}+11.02\%}$
test_redq_speed[True-backward] 13.5248ms 12.5319ms 79.7962 Ops/s 81.5053 Ops/s $\color{#d91a1a}-2.10\%$
test_redq_speed[reduce-overhead-None] 5.4371ms 4.6221ms 216.3518 Ops/s 215.8506 Ops/s $\color{#35bf28}+0.23\%$
test_redq_speed[reduce-overhead-backward] 13.3882ms 12.1833ms 82.0795 Ops/s 80.0449 Ops/s $\color{#35bf28}+2.54\%$
test_redq_deprec_speed[False-None] 15.3952ms 12.8311ms 77.9354 Ops/s 75.3128 Ops/s $\color{#35bf28}+3.48\%$
test_redq_deprec_speed[False-backward] 19.8523ms 18.6266ms 53.6866 Ops/s 52.2041 Ops/s $\color{#35bf28}+2.84\%$
test_redq_deprec_speed[True-None] 4.0209ms 3.5725ms 279.9127 Ops/s 265.4384 Ops/s $\textbf{\color{#35bf28}+5.45\%}$
test_redq_deprec_speed[True-backward] 8.2831ms 7.9686ms 125.4919 Ops/s 115.0623 Ops/s $\textbf{\color{#35bf28}+9.06\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.3267ms 3.6171ms 276.4670 Ops/s 266.4170 Ops/s $\color{#35bf28}+3.77\%$
test_redq_deprec_speed[reduce-overhead-backward] 9.7703ms 8.0591ms 124.0840 Ops/s 116.8564 Ops/s $\textbf{\color{#35bf28}+6.19\%}$
test_td3_speed[False-None] 33.3185ms 8.3277ms 120.0813 Ops/s 122.0472 Ops/s $\color{#d91a1a}-1.61\%$
test_td3_speed[False-backward] 10.9984ms 10.3903ms 96.2438 Ops/s 94.5083 Ops/s $\color{#35bf28}+1.84\%$
test_td3_speed[True-None] 2.7722ms 1.7394ms 574.9059 Ops/s 570.6132 Ops/s $\color{#35bf28}+0.75\%$
test_td3_speed[True-backward] 3.5787ms 3.3216ms 301.0562 Ops/s 293.6726 Ops/s $\color{#35bf28}+2.51\%$
test_td3_speed[reduce-overhead-None] 1.9503ms 1.7189ms 581.7828 Ops/s 569.6099 Ops/s $\color{#35bf28}+2.14\%$
test_td3_speed[reduce-overhead-backward] 3.3967ms 3.3131ms 301.8319 Ops/s 298.5917 Ops/s $\color{#35bf28}+1.09\%$
test_cql_speed[False-None] 39.5141ms 37.2113ms 26.8736 Ops/s 27.0605 Ops/s $\color{#d91a1a}-0.69\%$
test_cql_speed[False-backward] 49.5529ms 47.0232ms 21.2661 Ops/s 21.5285 Ops/s $\color{#d91a1a}-1.22\%$
test_cql_speed[True-None] 17.6425ms 15.8542ms 63.0746 Ops/s 64.2374 Ops/s $\color{#d91a1a}-1.81\%$
test_cql_speed[True-backward] 23.9773ms 22.2739ms 44.8956 Ops/s 44.6858 Ops/s $\color{#35bf28}+0.47\%$
test_cql_speed[reduce-overhead-None] 17.0911ms 15.7432ms 63.5195 Ops/s 63.1964 Ops/s $\color{#35bf28}+0.51\%$
test_cql_speed[reduce-overhead-backward] 23.9845ms 22.3027ms 44.8376 Ops/s 44.0790 Ops/s $\color{#35bf28}+1.72\%$
test_a2c_speed[False-None] 9.5584ms 7.4135ms 134.8885 Ops/s 139.1252 Ops/s $\color{#d91a1a}-3.05\%$
test_a2c_speed[False-backward] 15.0230ms 14.0884ms 70.9804 Ops/s 70.2751 Ops/s $\color{#35bf28}+1.00\%$
test_a2c_speed[True-None] 4.9857ms 4.2030ms 237.9270 Ops/s 235.0540 Ops/s $\color{#35bf28}+1.22\%$
test_a2c_speed[True-backward] 12.7332ms 10.7657ms 92.8878 Ops/s 93.5198 Ops/s $\color{#d91a1a}-0.68\%$
test_a2c_speed[reduce-overhead-None] 5.0532ms 4.2002ms 238.0866 Ops/s 234.8980 Ops/s $\color{#35bf28}+1.36\%$
test_a2c_speed[reduce-overhead-backward] 11.0132ms 10.6773ms 93.6563 Ops/s 90.2907 Ops/s $\color{#35bf28}+3.73\%$
test_ppo_speed[False-None] 7.9792ms 7.3843ms 135.4224 Ops/s 131.5160 Ops/s $\color{#35bf28}+2.97\%$
test_ppo_speed[False-backward] 16.3201ms 14.6817ms 68.1119 Ops/s 66.5112 Ops/s $\color{#35bf28}+2.41\%$
test_ppo_speed[True-None] 4.3027ms 3.6719ms 272.3419 Ops/s 268.7712 Ops/s $\color{#35bf28}+1.33\%$
test_ppo_speed[True-backward] 11.7294ms 9.6344ms 103.7946 Ops/s 103.7677 Ops/s $\color{#35bf28}+0.03\%$
test_ppo_speed[reduce-overhead-None] 4.7779ms 3.6893ms 271.0545 Ops/s 269.9039 Ops/s $\color{#35bf28}+0.43\%$
test_ppo_speed[reduce-overhead-backward] 11.9432ms 9.6116ms 104.0406 Ops/s 97.9956 Ops/s $\textbf{\color{#35bf28}+6.17\%}$
test_reinforce_speed[False-None] 7.7928ms 6.4780ms 154.3679 Ops/s 150.9115 Ops/s $\color{#35bf28}+2.29\%$
test_reinforce_speed[False-backward] 9.9679ms 9.7363ms 102.7081 Ops/s 102.2661 Ops/s $\color{#35bf28}+0.43\%$
test_reinforce_speed[True-None] 3.4647ms 2.6387ms 378.9721 Ops/s 364.2124 Ops/s $\color{#35bf28}+4.05\%$
test_reinforce_speed[True-backward] 9.1404ms 8.5983ms 116.3018 Ops/s 116.9116 Ops/s $\color{#d91a1a}-0.52\%$
test_reinforce_speed[reduce-overhead-None] 3.2311ms 2.6760ms 373.6894 Ops/s 369.8112 Ops/s $\color{#35bf28}+1.05\%$
test_reinforce_speed[reduce-overhead-backward] 9.3296ms 8.7536ms 114.2393 Ops/s 115.1671 Ops/s $\color{#d91a1a}-0.81\%$
test_iql_speed[False-None] 41.0834ms 34.2379ms 29.2074 Ops/s 30.4110 Ops/s $\color{#d91a1a}-3.96\%$
test_iql_speed[False-backward] 54.9182ms 46.6544ms 21.4342 Ops/s 21.5641 Ops/s $\color{#d91a1a}-0.60\%$
test_iql_speed[True-None] 12.2796ms 10.9649ms 91.2004 Ops/s 93.8016 Ops/s $\color{#d91a1a}-2.77\%$
test_iql_speed[True-backward] 23.4619ms 22.0634ms 45.3240 Ops/s 45.6755 Ops/s $\color{#d91a1a}-0.77\%$
test_iql_speed[reduce-overhead-None] 11.6704ms 10.8544ms 92.1289 Ops/s 93.4085 Ops/s $\color{#d91a1a}-1.37\%$
test_iql_speed[reduce-overhead-backward] 25.3319ms 22.1843ms 45.0770 Ops/s 45.9215 Ops/s $\color{#d91a1a}-1.84\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.1925ms 5.0021ms 199.9149 Ops/s 197.6127 Ops/s $\color{#35bf28}+1.17\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9174ms 0.5213ms 1.9183 KOps/s 1.8749 KOps/s $\color{#35bf28}+2.31\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9043ms 0.4998ms 2.0009 KOps/s 1.9971 KOps/s $\color{#35bf28}+0.19\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.2100ms 4.8512ms 206.1340 Ops/s 211.3587 Ops/s $\color{#d91a1a}-2.47\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.7763ms 0.5093ms 1.9635 KOps/s 1.9473 KOps/s $\color{#35bf28}+0.83\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9006ms 0.4902ms 2.0399 KOps/s 2.0387 KOps/s $\color{#35bf28}+0.06\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.3166ms 1.6321ms 612.7147 Ops/s 597.2123 Ops/s $\color{#35bf28}+2.60\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.2458ms 1.5778ms 633.7892 Ops/s 616.0092 Ops/s $\color{#35bf28}+2.89\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.2909ms 5.0811ms 196.8059 Ops/s 203.6805 Ops/s $\color{#d91a1a}-3.38\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.3066ms 0.6538ms 1.5294 KOps/s 1.5291 KOps/s $\color{#35bf28}+0.02\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9972ms 0.6282ms 1.5918 KOps/s 1.5857 KOps/s $\color{#35bf28}+0.39\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9846ms 4.9033ms 203.9437 Ops/s 208.7048 Ops/s $\color{#d91a1a}-2.28\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1225ms 0.5204ms 1.9214 KOps/s 1.8573 KOps/s $\color{#35bf28}+3.45\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8357ms 0.5023ms 1.9909 KOps/s 1.9957 KOps/s $\color{#d91a1a}-0.24\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.4683ms 4.8780ms 205.0026 Ops/s 211.2890 Ops/s $\color{#d91a1a}-2.98\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9425ms 0.5095ms 1.9626 KOps/s 1.9684 KOps/s $\color{#d91a1a}-0.29\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7452ms 0.4846ms 2.0638 KOps/s 2.0102 KOps/s $\color{#35bf28}+2.66\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.1910ms 4.9791ms 200.8403 Ops/s 200.4848 Ops/s $\color{#35bf28}+0.18\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.2113ms 0.6678ms 1.4975 KOps/s 1.5208 KOps/s $\color{#d91a1a}-1.53\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8281ms 0.6260ms 1.5974 KOps/s 1.5858 KOps/s $\color{#35bf28}+0.73\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.4580s 13.4393ms 74.4087 Ops/s 37.6119 Ops/s $\textbf{\color{#35bf28}+97.83\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.2286ms 2.1137ms 473.1100 Ops/s 428.0911 Ops/s $\textbf{\color{#35bf28}+10.52\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.2657ms 1.3282ms 752.8710 Ops/s 745.3070 Ops/s $\color{#35bf28}+1.01\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 5.8591ms 4.3764ms 228.4959 Ops/s 223.4155 Ops/s $\color{#35bf28}+2.27\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 6.5654ms 2.3102ms 432.8645 Ops/s 420.3026 Ops/s $\color{#35bf28}+2.99\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.7158ms 1.3977ms 715.4384 Ops/s 761.3456 Ops/s $\textbf{\color{#d91a1a}-6.03\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.3854s 12.2512ms 81.6248 Ops/s 226.8352 Ops/s $\textbf{\color{#d91a1a}-64.02\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.0606ms 2.5579ms 390.9429 Ops/s 403.7077 Ops/s $\color{#d91a1a}-3.16\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 5.2513ms 1.5421ms 648.4682 Ops/s 699.6142 Ops/s $\textbf{\color{#d91a1a}-7.31\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 15.4750ms 13.1381ms 76.1143 Ops/s 71.2265 Ops/s $\textbf{\color{#35bf28}+6.86\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.5778ms 15.2664ms 65.5033 Ops/s 64.5837 Ops/s $\color{#35bf28}+1.42\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 22.7818ms 21.6503ms 46.1887 Ops/s 43.8124 Ops/s $\textbf{\color{#35bf28}+5.42\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 17.0043ms 15.4021ms 64.9262 Ops/s 64.0622 Ops/s $\color{#35bf28}+1.35\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 23.6702ms 21.8776ms 45.7089 Ops/s 45.0721 Ops/s $\color{#35bf28}+1.41\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.0945ms 16.8672ms 59.2868 Ops/s 59.5821 Ops/s $\color{#d91a1a}-0.50\%$

@vmoens vmoens mentioned this pull request Dec 19, 2024
Merged
@github-actions GitHub Actions
Copy link

github-actions bot commented Dec 19, 2024 •
edited
Loading

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}21$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7305s 0.7214s 1.3863 Ops/s 1.3524 Ops/s $\color{#35bf28}+2.51\%$
test_transformed 0.9715s 0.9707s 1.0302 Ops/s 1.0304 Ops/s $\color{#d91a1a}-0.01\%$
test_serial 2.2375s 2.1571s 0.4636 Ops/s 0.4678 Ops/s $\color{#d91a1a}-0.90\%$
test_parallel 1.9445s 1.8717s 0.5343 Ops/s 0.5389 Ops/s $\color{#d91a1a}-0.85\%$
test_step_mdp_speed[True-True-True-True-True] 0.1488ms 39.5016μs 25.3154 KOps/s 25.6164 KOps/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-True-True-True-False] 0.2239ms 22.9853μs 43.5061 KOps/s 42.8202 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[True-True-True-False-True] 49.6720μs 22.1905μs 45.0642 KOps/s 45.7068 KOps/s $\color{#d91a1a}-1.41\%$
test_step_mdp_speed[True-True-True-False-False] 39.0320μs 13.1241μs 76.1959 KOps/s 78.3775 KOps/s $\color{#d91a1a}-2.78\%$
test_step_mdp_speed[True-True-False-True-True] 0.1052ms 42.9374μs 23.2897 KOps/s 23.4458 KOps/s $\color{#d91a1a}-0.67\%$
test_step_mdp_speed[True-True-False-True-False] 59.7730μs 25.6886μs 38.9278 KOps/s 38.9374 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[True-True-False-False-True] 52.9730μs 25.0255μs 39.9592 KOps/s 40.8633 KOps/s $\color{#d91a1a}-2.21\%$
test_step_mdp_speed[True-True-False-False-False] 45.1520μs 15.6973μs 63.7053 KOps/s 64.8442 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[True-False-True-True-True] 80.0440μs 44.9949μs 22.2247 KOps/s 22.1734 KOps/s $\color{#35bf28}+0.23\%$
test_step_mdp_speed[True-False-True-True-False] 85.7140μs 28.0889μs 35.6013 KOps/s 35.7706 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-False-True-False-True] 60.3230μs 24.9565μs 40.0697 KOps/s 40.5870 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-False-True-False-False] 41.3620μs 15.2928μs 65.3903 KOps/s 65.2854 KOps/s $\color{#35bf28}+0.16\%$
test_step_mdp_speed[True-False-False-True-True] 79.0340μs 46.4685μs 21.5200 KOps/s 21.1976 KOps/s $\color{#35bf28}+1.52\%$
test_step_mdp_speed[True-False-False-True-False] 58.5130μs 30.5883μs 32.6923 KOps/s 33.1554 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[True-False-False-False-True] 53.8230μs 27.0009μs 37.0359 KOps/s 37.3180 KOps/s $\color{#d91a1a}-0.76\%$
test_step_mdp_speed[True-False-False-False-False] 48.3530μs 17.5506μs 56.9781 KOps/s 56.7078 KOps/s $\color{#35bf28}+0.48\%$
test_step_mdp_speed[False-True-True-True-True] 71.5640μs 43.7121μs 22.8769 KOps/s 22.2449 KOps/s $\color{#35bf28}+2.84\%$
test_step_mdp_speed[False-True-True-True-False] 78.2040μs 27.7449μs 36.0427 KOps/s 35.7019 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[False-True-True-False-True] 0.1956ms 28.1409μs 35.5355 KOps/s 35.7216 KOps/s $\color{#d91a1a}-0.52\%$
test_step_mdp_speed[False-True-True-False-False] 0.1747ms 17.1992μs 58.1422 KOps/s 58.8796 KOps/s $\color{#d91a1a}-1.25\%$
test_step_mdp_speed[False-True-False-True-True] 74.3540μs 46.4797μs 21.5148 KOps/s 21.3977 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-True-False-True-False] 57.3430μs 29.9158μs 33.4272 KOps/s 33.1164 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[False-True-False-False-True] 3.1751ms 30.7684μs 32.5009 KOps/s 32.1728 KOps/s $\color{#35bf28}+1.02\%$
test_step_mdp_speed[False-True-False-False-False] 48.2520μs 19.6857μs 50.7983 KOps/s 52.2078 KOps/s $\color{#d91a1a}-2.70\%$
test_step_mdp_speed[False-False-True-True-True] 91.2140μs 49.3503μs 20.2633 KOps/s 20.7326 KOps/s $\color{#d91a1a}-2.26\%$
test_step_mdp_speed[False-False-True-True-False] 70.9340μs 32.9987μs 30.3042 KOps/s 31.4641 KOps/s $\color{#d91a1a}-3.69\%$
test_step_mdp_speed[False-False-True-False-True] 87.2440μs 30.0866μs 33.2373 KOps/s 33.0139 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[False-False-True-False-False] 48.5430μs 18.9360μs 52.8093 KOps/s 53.3454 KOps/s $\color{#d91a1a}-1.00\%$
test_step_mdp_speed[False-False-False-True-True] 80.1840μs 51.0218μs 19.5995 KOps/s 19.7638 KOps/s $\color{#d91a1a}-0.83\%$
test_step_mdp_speed[False-False-False-True-False] 60.8530μs 34.8148μs 28.7234 KOps/s 28.8473 KOps/s $\color{#d91a1a}-0.43\%$
test_step_mdp_speed[False-False-False-False-True] 65.9630μs 32.2759μs 30.9829 KOps/s 31.7856 KOps/s $\color{#d91a1a}-2.53\%$
test_step_mdp_speed[False-False-False-False-False] 85.0550μs 21.8382μs 45.7914 KOps/s 47.8139 KOps/s $\color{#d91a1a}-4.23\%$
test_values[generalized_advantage_estimate-True-True] 26.2900ms 25.7638ms 38.8141 Ops/s 39.4382 Ops/s $\color{#d91a1a}-1.58\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1169s 3.2447ms 308.1978 Ops/s 342.0200 Ops/s $\textbf{\color{#d91a1a}-9.89\%}$
test_values[td0_return_estimate-False-False] 0.1074ms 81.1466μs 12.3234 KOps/s 12.2940 KOps/s $\color{#35bf28}+0.24\%$
test_values[td1_return_estimate-False-False] 57.6419ms 57.1169ms 17.5079 Ops/s 17.7252 Ops/s $\color{#d91a1a}-1.23\%$
test_values[vec_td1_return_estimate-False-False] 1.2992ms 1.0946ms 913.5971 Ops/s 915.2903 Ops/s $\color{#d91a1a}-0.18\%$
test_values[td_lambda_return_estimate-True-False] 95.4811ms 90.7415ms 11.0203 Ops/s 11.2471 Ops/s $\color{#d91a1a}-2.02\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2605ms 1.0895ms 917.8452 Ops/s 918.5743 Ops/s $\color{#d91a1a}-0.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.0207ms 25.8364ms 38.7050 Ops/s 39.2623 Ops/s $\color{#d91a1a}-1.42\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0333ms 0.7649ms 1.3073 KOps/s 1.3241 KOps/s $\color{#d91a1a}-1.26\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7696ms 0.7115ms 1.4055 KOps/s 1.4787 KOps/s $\color{#d91a1a}-4.95\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6963ms 1.5029ms 665.3878 Ops/s 674.1815 Ops/s $\color{#d91a1a}-1.30\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8374ms 0.6978ms 1.4332 KOps/s 1.4420 KOps/s $\color{#d91a1a}-0.61\%$
test_dqn_speed[False-None] 6.9072ms 1.5479ms 646.0295 Ops/s 649.5570 Ops/s $\color{#d91a1a}-0.54\%$
test_dqn_speed[False-backward] 2.3361ms 2.1523ms 464.6267 Ops/s 465.7710 Ops/s $\color{#d91a1a}-0.25\%$
test_dqn_speed[True-None] 0.7141ms 0.5639ms 1.7734 KOps/s 1.7029 KOps/s $\color{#35bf28}+4.14\%$
test_dqn_speed[True-backward] 1.2788ms 1.1467ms 872.0821 Ops/s 786.7686 Ops/s $\textbf{\color{#35bf28}+10.84\%}$
test_dqn_speed[reduce-overhead-None] 0.9834ms 0.5740ms 1.7423 KOps/s 1.7172 KOps/s $\color{#35bf28}+1.46\%$
test_dqn_speed[reduce-overhead-backward] 1.1290ms 0.9909ms 1.0091 KOps/s 893.5637 Ops/s $\textbf{\color{#35bf28}+12.93\%}$
test_ddpg_speed[False-None] 3.2767ms 2.8922ms 345.7523 Ops/s 346.8476 Ops/s $\color{#d91a1a}-0.32\%$
test_ddpg_speed[False-backward] 4.6991ms 4.1790ms 239.2902 Ops/s 233.0394 Ops/s $\color{#35bf28}+2.68\%$
test_ddpg_speed[True-None] 1.5019ms 1.1082ms 902.3993 Ops/s 899.9342 Ops/s $\color{#35bf28}+0.27\%$
test_ddpg_speed[True-backward] 2.2976ms 2.2016ms 454.2108 Ops/s 415.7374 Ops/s $\textbf{\color{#35bf28}+9.25\%}$
test_ddpg_speed[reduce-overhead-None] 1.5011ms 1.1271ms 887.2578 Ops/s 892.6875 Ops/s $\color{#d91a1a}-0.61\%$
test_ddpg_speed[reduce-overhead-backward] 2.0788ms 1.6875ms 592.6072 Ops/s 534.2827 Ops/s $\textbf{\color{#35bf28}+10.92\%}$
test_sac_speed[False-None] 8.5620ms 8.1728ms 122.3576 Ops/s 121.9004 Ops/s $\color{#35bf28}+0.38\%$
test_sac_speed[False-backward] 11.6769ms 11.1189ms 89.9368 Ops/s 87.5953 Ops/s $\color{#35bf28}+2.67\%$
test_sac_speed[True-None] 1.8494ms 1.6063ms 622.5599 Ops/s 635.0924 Ops/s $\color{#d91a1a}-1.97\%$
test_sac_speed[True-backward] 3.5895ms 3.4659ms 288.5247 Ops/s 303.8883 Ops/s $\textbf{\color{#d91a1a}-5.06\%}$
test_sac_speed[reduce-overhead-None] 23.7463ms 12.8611ms 77.7537 Ops/s 78.3782 Ops/s $\color{#d91a1a}-0.80\%$
test_sac_speed[reduce-overhead-backward] 1.6788ms 1.5338ms 651.9605 Ops/s 725.9465 Ops/s $\textbf{\color{#d91a1a}-10.19\%}$
test_redq_speed[False-None] 8.3079ms 7.6073ms 131.4530 Ops/s 131.0820 Ops/s $\color{#35bf28}+0.28\%$
test_redq_speed[False-backward] 12.6138ms 11.7852ms 84.8520 Ops/s 87.2484 Ops/s $\color{#d91a1a}-2.75\%$
test_redq_speed[True-None] 2.3481ms 2.1227ms 471.1026 Ops/s 485.1436 Ops/s $\color{#d91a1a}-2.89\%$
test_redq_speed[True-backward] 4.1334ms 3.9484ms 253.2640 Ops/s 246.3783 Ops/s $\color{#35bf28}+2.79\%$
test_redq_speed[reduce-overhead-None] 2.4219ms 2.0382ms 490.6339 Ops/s 489.6018 Ops/s $\color{#35bf28}+0.21\%$
test_redq_speed[reduce-overhead-backward] 3.9796ms 3.9084ms 255.8606 Ops/s 255.5708 Ops/s $\color{#35bf28}+0.11\%$
test_redq_deprec_speed[False-None] 9.7608ms 9.2084ms 108.5959 Ops/s 107.3013 Ops/s $\color{#35bf28}+1.21\%$
test_redq_deprec_speed[False-backward] 13.0466ms 12.4541ms 80.2950 Ops/s 78.7782 Ops/s $\color{#35bf28}+1.93\%$
test_redq_deprec_speed[True-None] 2.7169ms 2.4695ms 404.9376 Ops/s 409.8075 Ops/s $\color{#d91a1a}-1.19\%$
test_redq_deprec_speed[True-backward] 4.6706ms 4.2685ms 234.2717 Ops/s 233.4149 Ops/s $\color{#35bf28}+0.37\%$
test_redq_deprec_speed[reduce-overhead-None] 2.5264ms 2.3704ms 421.8753 Ops/s 399.7065 Ops/s $\textbf{\color{#35bf28}+5.55\%}$
test_redq_deprec_speed[reduce-overhead-backward] 4.2540ms 4.1010ms 243.8422 Ops/s 244.9950 Ops/s $\color{#d91a1a}-0.47\%$
test_td3_speed[False-None] 8.1566ms 8.0502ms 124.2203 Ops/s 124.7432 Ops/s $\color{#d91a1a}-0.42\%$
test_td3_speed[False-backward] 11.1302ms 10.4833ms 95.3902 Ops/s 96.4290 Ops/s $\color{#d91a1a}-1.08\%$
test_td3_speed[True-None] 1.6684ms 1.6170ms 618.4344 Ops/s 614.9453 Ops/s $\color{#35bf28}+0.57\%$
test_td3_speed[True-backward] 3.3021ms 3.1725ms 315.2129 Ops/s 314.6365 Ops/s $\color{#35bf28}+0.18\%$
test_td3_speed[reduce-overhead-None] 82.4634ms 26.9096ms 37.1615 Ops/s 35.7096 Ops/s $\color{#35bf28}+4.07\%$
test_td3_speed[reduce-overhead-backward] 1.7353ms 1.3422ms 745.0250 Ops/s 744.7614 Ops/s $\color{#35bf28}+0.04\%$
test_cql_speed[False-None] 17.6229ms 17.0124ms 58.7805 Ops/s 58.3677 Ops/s $\color{#35bf28}+0.71\%$
test_cql_speed[False-backward] 22.8915ms 22.1816ms 45.0824 Ops/s 44.9506 Ops/s $\color{#35bf28}+0.29\%$
test_cql_speed[True-None] 3.2017ms 2.9988ms 333.4674 Ops/s 331.2559 Ops/s $\color{#35bf28}+0.67\%$
test_cql_speed[True-backward] 5.8196ms 5.4019ms 185.1210 Ops/s 189.7854 Ops/s $\color{#d91a1a}-2.46\%$
test_cql_speed[reduce-overhead-None] 21.9486ms 13.4123ms 74.5583 Ops/s 75.0118 Ops/s $\color{#d91a1a}-0.60\%$
test_cql_speed[reduce-overhead-backward] 1.7665ms 1.6310ms 613.1202 Ops/s 642.5058 Ops/s $\color{#d91a1a}-4.57\%$
test_a2c_speed[False-None] 3.4354ms 3.2649ms 306.2852 Ops/s 298.0680 Ops/s $\color{#35bf28}+2.76\%$
test_a2c_speed[False-backward] 6.6802ms 6.3341ms 157.8766 Ops/s 160.0582 Ops/s $\color{#d91a1a}-1.36\%$
test_a2c_speed[True-None] 1.4346ms 1.0355ms 965.6879 Ops/s 967.1012 Ops/s $\color{#d91a1a}-0.15\%$
test_a2c_speed[True-backward] 3.0640ms 2.6731ms 374.0909 Ops/s 355.9721 Ops/s $\textbf{\color{#35bf28}+5.09\%}$
test_a2c_speed[reduce-overhead-None] 22.2940ms 11.7479ms 85.1215 Ops/s 85.6077 Ops/s $\color{#d91a1a}-0.57\%$
test_a2c_speed[reduce-overhead-backward] 1.0496ms 0.9914ms 1.0086 KOps/s 995.6560 Ops/s $\color{#35bf28}+1.30\%$
test_ppo_speed[False-None] 4.1483ms 3.7536ms 266.4110 Ops/s 268.9464 Ops/s $\color{#d91a1a}-0.94\%$
test_ppo_speed[False-backward] 7.5344ms 7.0252ms 142.3446 Ops/s 145.1820 Ops/s $\color{#d91a1a}-1.95\%$
test_ppo_speed[True-None] 1.3702ms 0.9745ms 1.0262 KOps/s 1.0336 KOps/s $\color{#d91a1a}-0.71\%$
test_ppo_speed[True-backward] 2.7817ms 2.6010ms 384.4639 Ops/s 359.0732 Ops/s $\textbf{\color{#35bf28}+7.07\%}$
test_ppo_speed[reduce-overhead-None] 0.6934ms 0.5281ms 1.8936 KOps/s 1.8303 KOps/s $\color{#35bf28}+3.46\%$
test_ppo_speed[reduce-overhead-backward] 1.0377ms 0.9831ms 1.0172 KOps/s 858.5646 Ops/s $\textbf{\color{#35bf28}+18.47\%}$
test_reinforce_speed[False-None] 2.6740ms 2.3049ms 433.8524 Ops/s 435.5536 Ops/s $\color{#d91a1a}-0.39\%$
test_reinforce_speed[False-backward] 3.7081ms 3.3291ms 300.3825 Ops/s 289.6290 Ops/s $\color{#35bf28}+3.71\%$
test_reinforce_speed[True-None] 0.9981ms 0.8519ms 1.1738 KOps/s 1.1590 KOps/s $\color{#35bf28}+1.27\%$
test_reinforce_speed[True-backward] 2.8831ms 2.4985ms 400.2424 Ops/s 378.2384 Ops/s $\textbf{\color{#35bf28}+5.82\%}$
test_reinforce_speed[reduce-overhead-None] 20.0118ms 10.7806ms 92.7591 Ops/s 86.6457 Ops/s $\textbf{\color{#35bf28}+7.06\%}$
test_reinforce_speed[reduce-overhead-backward] 1.1892ms 1.0553ms 947.5829 Ops/s 806.2959 Ops/s $\textbf{\color{#35bf28}+17.52\%}$
test_iql_speed[False-None] 9.9375ms 9.4194ms 106.1644 Ops/s 105.1253 Ops/s $\color{#35bf28}+0.99\%$
test_iql_speed[False-backward] 13.6079ms 13.1071ms 76.2943 Ops/s 74.5647 Ops/s $\color{#35bf28}+2.32\%$
test_iql_speed[True-None] 2.2152ms 1.7967ms 556.5854 Ops/s 561.1887 Ops/s $\color{#d91a1a}-0.82\%$
test_iql_speed[True-backward] 4.5063ms 4.3485ms 229.9668 Ops/s 228.8345 Ops/s $\color{#35bf28}+0.49\%$
test_iql_speed[reduce-overhead-None] 20.5750ms 11.7408ms 85.1734 Ops/s 86.1155 Ops/s $\color{#d91a1a}-1.09\%$
test_iql_speed[reduce-overhead-backward] 1.5794ms 1.4591ms 685.3600 Ops/s 697.4926 Ops/s $\color{#d91a1a}-1.74\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0103ms 6.4457ms 155.1426 Ops/s 152.2440 Ops/s $\color{#35bf28}+1.90\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.4967ms 0.2760ms 3.6228 KOps/s 3.1852 KOps/s $\textbf{\color{#35bf28}+13.74\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6420ms 0.2549ms 3.9224 KOps/s 3.2452 KOps/s $\textbf{\color{#35bf28}+20.87\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.9689ms 6.1326ms 163.0619 Ops/s 159.5120 Ops/s $\color{#35bf28}+2.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.1285ms 0.3209ms 3.1166 KOps/s 3.3303 KOps/s $\textbf{\color{#d91a1a}-6.42\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4868ms 0.2718ms 3.6789 KOps/s 3.5732 KOps/s $\color{#35bf28}+2.96\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5858ms 1.2820ms 780.0029 Ops/s 805.8031 Ops/s $\color{#d91a1a}-3.20\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5721ms 1.3415ms 745.4090 Ops/s 841.8894 Ops/s $\textbf{\color{#d91a1a}-11.46\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5247ms 6.3395ms 157.7414 Ops/s 156.3447 Ops/s $\color{#35bf28}+0.89\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0434ms 0.4539ms 2.2031 KOps/s 2.4349 KOps/s $\textbf{\color{#d91a1a}-9.52\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7221ms 0.4301ms 2.3251 KOps/s 2.5554 KOps/s $\textbf{\color{#d91a1a}-9.01\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.4646ms 6.1925ms 161.4862 Ops/s 160.6395 Ops/s $\color{#35bf28}+0.53\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0085ms 0.3305ms 3.0258 KOps/s 3.1646 KOps/s $\color{#d91a1a}-4.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4942ms 0.2950ms 3.3902 KOps/s 3.5006 KOps/s $\color{#d91a1a}-3.15\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.5407ms 6.1478ms 162.6603 Ops/s 160.5078 Ops/s $\color{#35bf28}+1.34\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7737ms 0.2922ms 3.4219 KOps/s 3.0655 KOps/s $\textbf{\color{#35bf28}+11.63\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4772ms 0.2638ms 3.7901 KOps/s 3.1791 KOps/s $\textbf{\color{#35bf28}+19.22\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6305ms 6.3512ms 157.4514 Ops/s 155.8128 Ops/s $\color{#35bf28}+1.05\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9179ms 0.4360ms 2.2938 KOps/s 2.2178 KOps/s $\color{#35bf28}+3.43\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6094ms 0.3911ms 2.5566 KOps/s 2.2578 KOps/s $\textbf{\color{#35bf28}+13.23\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1292ms 5.4430ms 183.7219 Ops/s 182.9230 Ops/s $\color{#35bf28}+0.44\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.0612ms 1.7633ms 567.1149 Ops/s 442.0223 Ops/s $\textbf{\color{#35bf28}+28.30\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.3284ms 1.2684ms 788.3652 Ops/s 816.3402 Ops/s $\color{#d91a1a}-3.43\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.5233ms 5.5399ms 180.5096 Ops/s 183.5876 Ops/s $\color{#d91a1a}-1.68\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.0778ms 2.0337ms 491.7186 Ops/s 435.1271 Ops/s $\textbf{\color{#35bf28}+13.01\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.0860ms 0.9431ms 1.0603 KOps/s 849.4236 Ops/s $\textbf{\color{#35bf28}+24.83\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5057s 15.7465ms 63.5061 Ops/s 32.7292 Ops/s $\textbf{\color{#35bf28}+94.03\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 10.7624ms 2.2471ms 445.0168 Ops/s 432.8503 Ops/s $\color{#35bf28}+2.81\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.2853ms 1.3294ms 752.2027 Ops/s 708.7617 Ops/s $\textbf{\color{#35bf28}+6.13\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 16.0735ms 15.5488ms 64.3137 Ops/s 64.6991 Ops/s $\color{#d91a1a}-0.60\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.5661ms 17.4444ms 57.3251 Ops/s 55.0005 Ops/s $\color{#35bf28}+4.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 20.8713ms 19.8048ms 50.4928 Ops/s 49.2286 Ops/s $\color{#35bf28}+2.57\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.4497ms 18.0112ms 55.5209 Ops/s 55.7523 Ops/s $\color{#d91a1a}-0.42\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.3947ms 19.9951ms 50.0123 Ops/s 49.0671 Ops/s $\color{#35bf28}+1.93\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.6310ms 19.7469ms 50.6407 Ops/s 51.0028 Ops/s $\color{#d91a1a}-0.71\%$

@vmoens vmoens mentioned this pull request Dec 19, 2024
Merged
@vmoens vmoens added the bug Something isn't working label Dec 20, 2024
@vmoens
Update
9a60ec9 
[ghstack-poisoned]
@vmoens vmoens mentioned this pull request Dec 20, 2024
Merged
@vmoens vmoens merged commit 9a60ec9 into gh/vmoens/60/base Dec 20, 2024
64 of 79 checks passed
vmoens added a commit that referenced this pull request Dec 20, 2024
@vmoens
[BugFix] Avoid KeyError in slice sampler (for compile)
21eeca4 
ghstack-source-id: 6e2a3036f0e50d365387cced50a761b97a47317d
Pull Request resolved: #2670
@vmoens vmoens deleted the gh/vmoens/60/head branch December 20, 2024 10:27
@vmoens vmoens mentioned this pull request Dec 20, 2024
Closed
3 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
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
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vmoens @facebook-github-bot