Skip to content

[Feature] Env with tensorclass attributes #2788

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

Merged
merged 5 commits into from
Feb 13, 2025
Merged

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Feb 13, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2788

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 3730096 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.

vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: 1d844e10ecfa67a4e1052bb865cd0f529c860692
Pull Request resolved: #2788
@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 13, 2025
Copy link

github-actions bot commented Feb 13, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.5909s 0.5034s 1.9864 Ops/s 2.0108 Ops/s $\color{#d91a1a}-1.21\%$
test_transformed 1.0709s 0.9819s 1.0185 Ops/s 1.0294 Ops/s $\color{#d91a1a}-1.07\%$
test_serial 1.5668s 1.4773s 0.6769 Ops/s 0.6746 Ops/s $\color{#35bf28}+0.35\%$
test_parallel 1.3609s 1.2722s 0.7860 Ops/s 0.7796 Ops/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-True-True-True-True] 0.1980ms 29.8697μs 33.4788 KOps/s 33.2496 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[True-True-True-True-False] 42.3290μs 17.8418μs 56.0482 KOps/s 56.6066 KOps/s $\color{#d91a1a}-0.99\%$
test_step_mdp_speed[True-True-True-False-True] 50.7450μs 16.9961μs 58.8369 KOps/s 58.9081 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-True-True-False-False] 30.1860μs 10.0083μs 99.9175 KOps/s 100.8284 KOps/s $\color{#d91a1a}-0.90\%$
test_step_mdp_speed[True-True-False-True-True] 65.5930μs 31.9002μs 31.3478 KOps/s 31.3397 KOps/s $\color{#35bf28}+0.03\%$
test_step_mdp_speed[True-True-False-True-False] 44.2130μs 19.5248μs 51.2168 KOps/s 51.3170 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[True-True-False-False-True] 45.2140μs 18.8405μs 53.0771 KOps/s 52.9919 KOps/s $\color{#35bf28}+0.16\%$
test_step_mdp_speed[True-True-False-False-False] 32.8120μs 11.8324μs 84.5140 KOps/s 85.0294 KOps/s $\color{#d91a1a}-0.61\%$
test_step_mdp_speed[True-False-True-True-True] 97.2620μs 34.1046μs 29.3215 KOps/s 29.4235 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-False-True-True-False] 48.0100μs 21.5088μs 46.4926 KOps/s 46.5781 KOps/s $\color{#d91a1a}-0.18\%$
test_step_mdp_speed[True-False-True-False-True] 0.5072ms 18.8689μs 52.9973 KOps/s 52.6237 KOps/s $\color{#35bf28}+0.71\%$
test_step_mdp_speed[True-False-True-False-False] 50.6950μs 11.7807μs 84.8844 KOps/s 84.0052 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-False-False-True-True] 77.8350μs 35.2483μs 28.3701 KOps/s 28.4149 KOps/s $\color{#d91a1a}-0.16\%$
test_step_mdp_speed[True-False-False-True-False] 50.1830μs 23.0757μs 43.3356 KOps/s 43.6704 KOps/s $\color{#d91a1a}-0.77\%$
test_step_mdp_speed[True-False-False-False-True] 46.5560μs 20.5000μs 48.7804 KOps/s 48.6958 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-False-False-False-False] 38.8630μs 13.5595μs 73.7492 KOps/s 72.7403 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[False-True-True-True-True] 73.9890μs 33.5988μs 29.7630 KOps/s 30.1852 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[False-True-True-True-False] 46.2760μs 21.3741μs 46.7856 KOps/s 47.2649 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-True-True-False-True] 50.2640μs 21.3436μs 46.8525 KOps/s 46.8683 KOps/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[False-True-True-False-False] 37.1400μs 13.2104μs 75.6981 KOps/s 75.4010 KOps/s $\color{#35bf28}+0.39\%$
test_step_mdp_speed[False-True-False-True-True] 76.6930μs 35.3969μs 28.2511 KOps/s 28.4722 KOps/s $\color{#d91a1a}-0.78\%$
test_step_mdp_speed[False-True-False-True-False] 61.1340μs 23.1732μs 43.1533 KOps/s 43.8998 KOps/s $\color{#d91a1a}-1.70\%$
test_step_mdp_speed[False-True-False-False-True] 2.8245ms 23.0310μs 43.4197 KOps/s 43.1566 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[False-True-False-False-False] 40.0240μs 14.9333μs 66.9642 KOps/s 67.5233 KOps/s $\color{#d91a1a}-0.83\%$
test_step_mdp_speed[False-False-True-True-True] 73.2570μs 37.2386μs 26.8539 KOps/s 27.1984 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[False-False-True-True-False] 56.0350μs 25.0038μs 39.9939 KOps/s 40.6257 KOps/s $\color{#d91a1a}-1.56\%$
test_step_mdp_speed[False-False-True-False-True] 57.3970μs 23.0577μs 43.3694 KOps/s 43.5072 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[False-False-True-False-False] 0.5950ms 14.8306μs 67.4280 KOps/s 67.9662 KOps/s $\color{#d91a1a}-0.79\%$
test_step_mdp_speed[False-False-False-True-True] 77.6650μs 38.6635μs 25.8642 KOps/s 25.9990 KOps/s $\color{#d91a1a}-0.52\%$
test_step_mdp_speed[False-False-False-True-False] 66.7950μs 26.6694μs 37.4962 KOps/s 37.9054 KOps/s $\color{#d91a1a}-1.08\%$
test_step_mdp_speed[False-False-False-False-True] 0.2776ms 25.4021μs 39.3669 KOps/s 40.7242 KOps/s $\color{#d91a1a}-3.33\%$
test_step_mdp_speed[False-False-False-False-False] 42.7600μs 16.7017μs 59.8740 KOps/s 60.5942 KOps/s $\color{#d91a1a}-1.19\%$
test_values[generalized_advantage_estimate-True-True] 15.2950ms 10.3352ms 96.7566 Ops/s 100.0079 Ops/s $\color{#d91a1a}-3.25\%$
test_values[vec_generalized_advantage_estimate-True-True] 26.8065ms 24.2495ms 41.2379 Ops/s 41.3647 Ops/s $\color{#d91a1a}-0.31\%$
test_values[td0_return_estimate-False-False] 0.2392ms 0.1928ms 5.1862 KOps/s 5.6393 KOps/s $\textbf{\color{#d91a1a}-8.04\%}$
test_values[td1_return_estimate-False-False] 26.4110ms 23.9096ms 41.8242 Ops/s 41.1468 Ops/s $\color{#35bf28}+1.65\%$
test_values[vec_td1_return_estimate-False-False] 30.1346ms 24.4236ms 40.9440 Ops/s 41.1704 Ops/s $\color{#d91a1a}-0.55\%$
test_values[td_lambda_return_estimate-True-False] 37.6196ms 34.3530ms 29.1096 Ops/s 28.3994 Ops/s $\color{#35bf28}+2.50\%$
test_values[vec_td_lambda_return_estimate-True-False] 25.7730ms 24.2407ms 41.2529 Ops/s 40.7715 Ops/s $\color{#35bf28}+1.18\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.0407ms 8.4455ms 118.4068 Ops/s 117.8749 Ops/s $\color{#35bf28}+0.45\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4197ms 1.8044ms 554.2053 Ops/s 519.6007 Ops/s $\textbf{\color{#35bf28}+6.66\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4599ms 0.3653ms 2.7378 KOps/s 2.7753 KOps/s $\color{#d91a1a}-1.35\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 55.4603ms 38.7091ms 25.8337 Ops/s 25.1182 Ops/s $\color{#35bf28}+2.85\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.1914ms 3.4570ms 289.2687 Ops/s 291.1144 Ops/s $\color{#d91a1a}-0.63\%$
test_dqn_speed[False-None] 6.4483ms 1.3925ms 718.1231 Ops/s 711.4536 Ops/s $\color{#35bf28}+0.94\%$
test_dqn_speed[False-backward] 1.9458ms 1.8818ms 531.4135 Ops/s 534.6242 Ops/s $\color{#d91a1a}-0.60\%$
test_dqn_speed[True-None] 0.5910ms 0.4919ms 2.0328 KOps/s 1.9863 KOps/s $\color{#35bf28}+2.34\%$
test_dqn_speed[True-backward] 0.9890ms 0.9321ms 1.0729 KOps/s 1.1010 KOps/s $\color{#d91a1a}-2.55\%$
test_dqn_speed[reduce-overhead-None] 0.7297ms 0.4925ms 2.0304 KOps/s 1.9471 KOps/s $\color{#35bf28}+4.28\%$
test_dqn_speed[reduce-overhead-backward] 1.0807ms 0.9261ms 1.0798 KOps/s 1.0768 KOps/s $\color{#35bf28}+0.28\%$
test_ddpg_speed[False-None] 3.2162ms 2.8778ms 347.4845 Ops/s 345.3992 Ops/s $\color{#35bf28}+0.60\%$
test_ddpg_speed[False-backward] 4.1569ms 4.0611ms 246.2387 Ops/s 250.3745 Ops/s $\color{#d91a1a}-1.65\%$
test_ddpg_speed[True-None] 1.5975ms 1.2416ms 805.3818 Ops/s 801.2517 Ops/s $\color{#35bf28}+0.52\%$
test_ddpg_speed[True-backward] 2.2604ms 2.1557ms 463.8807 Ops/s 443.9544 Ops/s $\color{#35bf28}+4.49\%$
test_ddpg_speed[reduce-overhead-None] 3.5722ms 1.2846ms 778.4627 Ops/s 797.5163 Ops/s $\color{#d91a1a}-2.39\%$
test_ddpg_speed[reduce-overhead-backward] 2.1925ms 2.1406ms 467.1669 Ops/s 464.6341 Ops/s $\color{#35bf28}+0.55\%$
test_sac_speed[False-None] 9.1873ms 8.0409ms 124.3640 Ops/s 119.2968 Ops/s $\color{#35bf28}+4.25\%$
test_sac_speed[False-backward] 12.9420ms 10.8120ms 92.4898 Ops/s 93.3554 Ops/s $\color{#d91a1a}-0.93\%$
test_sac_speed[True-None] 2.5758ms 2.1336ms 468.6888 Ops/s 464.9167 Ops/s $\color{#35bf28}+0.81\%$
test_sac_speed[True-backward] 4.7803ms 3.8782ms 257.8541 Ops/s 244.8663 Ops/s $\textbf{\color{#35bf28}+5.30\%}$
test_sac_speed[reduce-overhead-None] 2.7523ms 2.1298ms 469.5264 Ops/s 460.1081 Ops/s $\color{#35bf28}+2.05\%$
test_sac_speed[reduce-overhead-backward] 3.9169ms 3.8351ms 260.7471 Ops/s 252.2484 Ops/s $\color{#35bf28}+3.37\%$
test_redq_speed[False-None] 14.8691ms 12.8449ms 77.8522 Ops/s 51.5985 Ops/s $\textbf{\color{#35bf28}+50.88\%}$
test_redq_speed[False-backward] 23.4005ms 22.0569ms 45.3373 Ops/s 44.2395 Ops/s $\color{#35bf28}+2.48\%$
test_redq_speed[True-None] 5.8260ms 4.9482ms 202.0925 Ops/s 182.0174 Ops/s $\textbf{\color{#35bf28}+11.03\%}$
test_redq_speed[True-backward] 13.6278ms 12.2800ms 81.4330 Ops/s 76.3238 Ops/s $\textbf{\color{#35bf28}+6.69\%}$
test_redq_speed[reduce-overhead-None] 6.0267ms 4.8912ms 204.4501 Ops/s 206.4085 Ops/s $\color{#d91a1a}-0.95\%$
test_redq_speed[reduce-overhead-backward] 12.6789ms 12.2866ms 81.3896 Ops/s 80.6593 Ops/s $\color{#35bf28}+0.91\%$
test_redq_deprec_speed[False-None] 25.2705ms 13.0127ms 76.8482 Ops/s 79.3099 Ops/s $\color{#d91a1a}-3.10\%$
test_redq_deprec_speed[False-backward] 19.4571ms 18.1871ms 54.9841 Ops/s 55.6760 Ops/s $\color{#d91a1a}-1.24\%$
test_redq_deprec_speed[True-None] 4.2600ms 3.8377ms 260.5695 Ops/s 259.6086 Ops/s $\color{#35bf28}+0.37\%$
test_redq_deprec_speed[True-backward] 8.4860ms 8.2127ms 121.7629 Ops/s 120.9791 Ops/s $\color{#35bf28}+0.65\%$
test_redq_deprec_speed[reduce-overhead-None] 4.6699ms 3.8397ms 260.4384 Ops/s 253.5343 Ops/s $\color{#35bf28}+2.72\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.6041ms 8.2039ms 121.8938 Ops/s 113.1342 Ops/s $\textbf{\color{#35bf28}+7.74\%}$
test_td3_speed[False-None] 9.6984ms 7.9873ms 125.1983 Ops/s 123.9397 Ops/s $\color{#35bf28}+1.02\%$
test_td3_speed[False-backward] 11.5911ms 10.4740ms 95.4742 Ops/s 94.5641 Ops/s $\color{#35bf28}+0.96\%$
test_td3_speed[True-None] 2.0696ms 1.8182ms 550.0007 Ops/s 529.7019 Ops/s $\color{#35bf28}+3.83\%$
test_td3_speed[True-backward] 3.4640ms 3.4171ms 292.6486 Ops/s 283.3916 Ops/s $\color{#35bf28}+3.27\%$
test_td3_speed[reduce-overhead-None] 2.0257ms 1.8104ms 552.3533 Ops/s 534.6493 Ops/s $\color{#35bf28}+3.31\%$
test_td3_speed[reduce-overhead-backward] 4.3802ms 3.4647ms 288.6292 Ops/s 289.7672 Ops/s $\color{#d91a1a}-0.39\%$
test_cql_speed[False-None] 39.0692ms 35.9976ms 27.7796 Ops/s 27.0964 Ops/s $\color{#35bf28}+2.52\%$
test_cql_speed[False-backward] 53.3119ms 46.1425ms 21.6720 Ops/s 21.4149 Ops/s $\color{#35bf28}+1.20\%$
test_cql_speed[True-None] 17.7505ms 15.9555ms 62.6744 Ops/s 61.1108 Ops/s $\color{#35bf28}+2.56\%$
test_cql_speed[True-backward] 23.3387ms 22.6508ms 44.1486 Ops/s 44.7168 Ops/s $\color{#d91a1a}-1.27\%$
test_cql_speed[reduce-overhead-None] 17.4597ms 16.2303ms 61.6133 Ops/s 62.7752 Ops/s $\color{#d91a1a}-1.85\%$
test_cql_speed[reduce-overhead-backward] 24.2150ms 23.0610ms 43.3633 Ops/s 44.5852 Ops/s $\color{#d91a1a}-2.74\%$
test_a2c_speed[False-None] 8.5050ms 7.1593ms 139.6785 Ops/s 138.9208 Ops/s $\color{#35bf28}+0.55\%$
test_a2c_speed[False-backward] 14.9015ms 14.4641ms 69.1367 Ops/s 70.5645 Ops/s $\color{#d91a1a}-2.02\%$
test_a2c_speed[True-None] 3.9371ms 3.7239ms 268.5344 Ops/s 264.2308 Ops/s $\color{#35bf28}+1.63\%$
test_a2c_speed[True-backward] 11.5873ms 10.4803ms 95.4170 Ops/s 99.1481 Ops/s $\color{#d91a1a}-3.76\%$
test_a2c_speed[reduce-overhead-None] 4.4571ms 3.7738ms 264.9883 Ops/s 263.8196 Ops/s $\color{#35bf28}+0.44\%$
test_a2c_speed[reduce-overhead-backward] 11.2689ms 10.4544ms 95.6537 Ops/s 99.3944 Ops/s $\color{#d91a1a}-3.76\%$
test_ppo_speed[False-None] 8.8939ms 7.6819ms 130.1761 Ops/s 133.7821 Ops/s $\color{#d91a1a}-2.70\%$
test_ppo_speed[False-backward] 15.9233ms 15.3503ms 65.1455 Ops/s 68.8780 Ops/s $\textbf{\color{#d91a1a}-5.42\%}$
test_ppo_speed[True-None] 4.9155ms 4.1346ms 241.8632 Ops/s 240.9256 Ops/s $\color{#35bf28}+0.39\%$
test_ppo_speed[True-backward] 11.2105ms 10.0681ms 99.3236 Ops/s 100.7221 Ops/s $\color{#d91a1a}-1.39\%$
test_ppo_speed[reduce-overhead-None] 4.6174ms 4.1781ms 239.3407 Ops/s 239.3135 Ops/s $\color{#35bf28}+0.01\%$
test_ppo_speed[reduce-overhead-backward] 12.0662ms 10.8757ms 91.9483 Ops/s 100.2393 Ops/s $\textbf{\color{#d91a1a}-8.27\%}$
test_reinforce_speed[False-None] 8.0215ms 6.7988ms 147.0857 Ops/s 153.0135 Ops/s $\color{#d91a1a}-3.87\%$
test_reinforce_speed[False-backward] 10.5744ms 10.2223ms 97.8251 Ops/s 103.3376 Ops/s $\textbf{\color{#d91a1a}-5.33\%}$
test_reinforce_speed[True-None] 4.0321ms 3.2319ms 309.4134 Ops/s 318.2356 Ops/s $\color{#d91a1a}-2.77\%$
test_reinforce_speed[True-backward] 10.4909ms 9.1654ms 109.1060 Ops/s 111.0185 Ops/s $\color{#d91a1a}-1.72\%$
test_reinforce_speed[reduce-overhead-None] 4.6948ms 3.0865ms 323.9956 Ops/s 306.7556 Ops/s $\textbf{\color{#35bf28}+5.62\%}$
test_reinforce_speed[reduce-overhead-backward] 12.5840ms 9.9901ms 100.0987 Ops/s 111.2811 Ops/s $\textbf{\color{#d91a1a}-10.05\%}$
test_iql_speed[False-None] 40.8089ms 34.7341ms 28.7902 Ops/s 30.3580 Ops/s $\textbf{\color{#d91a1a}-5.16\%}$
test_iql_speed[False-backward] 47.9032ms 46.4561ms 21.5257 Ops/s 21.7921 Ops/s $\color{#d91a1a}-1.22\%$
test_iql_speed[True-None] 12.9484ms 11.5128ms 86.8595 Ops/s 88.2613 Ops/s $\color{#d91a1a}-1.59\%$
test_iql_speed[True-backward] 24.1309ms 22.9088ms 43.6513 Ops/s 45.2517 Ops/s $\color{#d91a1a}-3.54\%$
test_iql_speed[reduce-overhead-None] 12.3247ms 11.6015ms 86.1959 Ops/s 87.9173 Ops/s $\color{#d91a1a}-1.96\%$
test_iql_speed[reduce-overhead-backward] 24.5646ms 23.2937ms 42.9300 Ops/s 44.5619 Ops/s $\color{#d91a1a}-3.66\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.9839ms 5.2102ms 191.9295 Ops/s 206.1131 Ops/s $\textbf{\color{#d91a1a}-6.88\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9754ms 0.5620ms 1.7794 KOps/s 1.9500 KOps/s $\textbf{\color{#d91a1a}-8.75\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7387ms 0.5028ms 1.9888 KOps/s 2.0296 KOps/s $\color{#d91a1a}-2.01\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.7353ms 4.8982ms 204.1551 Ops/s 214.8313 Ops/s $\color{#d91a1a}-4.97\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.3065ms 0.5159ms 1.9383 KOps/s 1.9764 KOps/s $\color{#d91a1a}-1.93\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8259ms 0.4947ms 2.0215 KOps/s 2.0944 KOps/s $\color{#d91a1a}-3.48\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.9895ms 1.6533ms 604.8537 Ops/s 611.0322 Ops/s $\color{#d91a1a}-1.01\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.1575ms 1.5740ms 635.3098 Ops/s 643.5122 Ops/s $\color{#d91a1a}-1.27\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.2013ms 5.0237ms 199.0578 Ops/s 210.3631 Ops/s $\textbf{\color{#d91a1a}-5.37\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5060ms 0.6558ms 1.5248 KOps/s 1.5292 KOps/s $\color{#d91a1a}-0.29\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8935ms 0.6322ms 1.5817 KOps/s 1.6176 KOps/s $\color{#d91a1a}-2.22\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.4035ms 4.8455ms 206.3769 Ops/s 217.2190 Ops/s $\color{#d91a1a}-4.99\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8822ms 0.5204ms 1.9217 KOps/s 1.9243 KOps/s $\color{#d91a1a}-0.14\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7834ms 0.5043ms 1.9829 KOps/s 2.0663 KOps/s $\color{#d91a1a}-4.04\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.1476ms 4.8169ms 207.6029 Ops/s 216.7641 Ops/s $\color{#d91a1a}-4.23\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9167ms 0.5240ms 1.9083 KOps/s 1.9888 KOps/s $\color{#d91a1a}-4.05\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7730ms 0.4911ms 2.0363 KOps/s 2.0950 KOps/s $\color{#d91a1a}-2.81\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.9836ms 4.9995ms 200.0210 Ops/s 210.4575 Ops/s $\color{#d91a1a}-4.96\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1348ms 0.6559ms 1.5247 KOps/s 1.5542 KOps/s $\color{#d91a1a}-1.90\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9407ms 0.6392ms 1.5645 KOps/s 1.6022 KOps/s $\color{#d91a1a}-2.35\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.7683ms 4.2718ms 234.0918 Ops/s 240.9016 Ops/s $\color{#d91a1a}-2.83\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 9.0843ms 2.3823ms 419.7637 Ops/s 438.8520 Ops/s $\color{#d91a1a}-4.35\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.6354ms 1.4422ms 693.3627 Ops/s 717.6780 Ops/s $\color{#d91a1a}-3.39\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 5.7208ms 4.2735ms 233.9987 Ops/s 247.0548 Ops/s $\textbf{\color{#d91a1a}-5.28\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.4945s 12.1879ms 82.0485 Ops/s 386.1012 Ops/s $\textbf{\color{#d91a1a}-78.75\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.5875ms 1.5163ms 659.5018 Ops/s 706.0278 Ops/s $\textbf{\color{#d91a1a}-6.59\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.1594ms 4.5179ms 221.3415 Ops/s 33.2805 Ops/s $\textbf{\color{#35bf28}+565.08\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 7.2269ms 2.5490ms 392.3126 Ops/s 403.4104 Ops/s $\color{#d91a1a}-2.75\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 12.5751ms 1.7818ms 561.2403 Ops/s 679.2541 Ops/s $\textbf{\color{#d91a1a}-17.37\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.4688ms 11.8156ms 84.6339 Ops/s 80.6599 Ops/s $\color{#35bf28}+4.93\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 16.6426ms 14.6109ms 68.4422 Ops/s 69.1155 Ops/s $\color{#d91a1a}-0.97\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 21.8610ms 20.7725ms 48.1407 Ops/s 46.7254 Ops/s $\color{#35bf28}+3.03\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.7196ms 14.8338ms 67.4135 Ops/s 67.6831 Ops/s $\color{#d91a1a}-0.40\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 22.0402ms 20.7605ms 48.1685 Ops/s 48.0496 Ops/s $\color{#35bf28}+0.25\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 17.0140ms 15.9405ms 62.7333 Ops/s 63.1279 Ops/s $\color{#d91a1a}-0.63\%$

Copy link

github-actions bot commented Feb 13, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.9039s 0.8157s 1.2260 Ops/s 1.2046 Ops/s $\color{#35bf28}+1.77\%$
test_transformed 1.4824s 1.3996s 0.7145 Ops/s 0.6923 Ops/s $\color{#35bf28}+3.22\%$
test_serial 2.3890s 2.3059s 0.4337 Ops/s 0.4252 Ops/s $\color{#35bf28}+2.00\%$
test_parallel 1.9685s 1.8658s 0.5360 Ops/s 0.5385 Ops/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-True-True-True-True] 0.1827ms 41.5271μs 24.0807 KOps/s 24.9650 KOps/s $\color{#d91a1a}-3.54\%$
test_step_mdp_speed[True-True-True-True-False] 55.1310μs 23.8136μs 41.9927 KOps/s 41.9758 KOps/s $\color{#35bf28}+0.04\%$
test_step_mdp_speed[True-True-True-False-True] 77.0910μs 22.4035μs 44.6358 KOps/s 44.9932 KOps/s $\color{#d91a1a}-0.79\%$
test_step_mdp_speed[True-True-True-False-False] 47.9600μs 13.2828μs 75.2854 KOps/s 76.6346 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[True-True-False-True-True] 88.3320μs 43.4298μs 23.0257 KOps/s 23.0739 KOps/s $\color{#d91a1a}-0.21\%$
test_step_mdp_speed[True-True-False-True-False] 68.8210μs 26.0847μs 38.3367 KOps/s 38.4500 KOps/s $\color{#d91a1a}-0.29\%$
test_step_mdp_speed[True-True-False-False-True] 62.8210μs 25.2613μs 39.5862 KOps/s 39.6348 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[True-True-False-False-False] 56.5800μs 15.6409μs 63.9349 KOps/s 63.5012 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[True-False-True-True-True] 79.0910μs 45.9914μs 21.7432 KOps/s 21.8303 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[True-False-True-True-False] 64.1410μs 28.5451μs 35.0323 KOps/s 34.7538 KOps/s $\color{#35bf28}+0.80\%$
test_step_mdp_speed[True-False-True-False-True] 71.1210μs 25.1429μs 39.7727 KOps/s 40.5836 KOps/s $\color{#d91a1a}-2.00\%$
test_step_mdp_speed[True-False-True-False-False] 47.5810μs 15.5233μs 64.4195 KOps/s 64.1487 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[True-False-False-True-True] 87.9420μs 47.8706μs 20.8896 KOps/s 20.9404 KOps/s $\color{#d91a1a}-0.24\%$
test_step_mdp_speed[True-False-False-True-False] 63.9710μs 30.7842μs 32.4842 KOps/s 32.6039 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-False-False-False-True] 65.7210μs 27.5572μs 36.2882 KOps/s 37.3909 KOps/s $\color{#d91a1a}-2.95\%$
test_step_mdp_speed[True-False-False-False-False] 47.4510μs 17.7785μs 56.2478 KOps/s 56.3707 KOps/s $\color{#d91a1a}-0.22\%$
test_step_mdp_speed[False-True-True-True-True] 87.0110μs 45.7566μs 21.8548 KOps/s 22.1088 KOps/s $\color{#d91a1a}-1.15\%$
test_step_mdp_speed[False-True-True-True-False] 58.6910μs 28.5345μs 35.0452 KOps/s 35.3891 KOps/s $\color{#d91a1a}-0.97\%$
test_step_mdp_speed[False-True-True-False-True] 72.4910μs 28.5780μs 34.9919 KOps/s 35.0263 KOps/s $\color{#d91a1a}-0.10\%$
test_step_mdp_speed[False-True-True-False-False] 49.2500μs 17.4738μs 57.2284 KOps/s 58.0665 KOps/s $\color{#d91a1a}-1.44\%$
test_step_mdp_speed[False-True-False-True-True] 94.6110μs 48.4730μs 20.6300 KOps/s 21.3462 KOps/s $\color{#d91a1a}-3.36\%$
test_step_mdp_speed[False-True-False-True-False] 72.2210μs 30.8705μs 32.3934 KOps/s 32.7591 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[False-True-False-False-True] 3.1411ms 31.2024μs 32.0489 KOps/s 31.5610 KOps/s $\color{#35bf28}+1.55\%$
test_step_mdp_speed[False-True-False-False-False] 51.6410μs 19.5425μs 51.1705 KOps/s 51.0212 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[False-False-True-True-True] 94.9110μs 50.3480μs 19.8618 KOps/s 20.0438 KOps/s $\color{#d91a1a}-0.91\%$
test_step_mdp_speed[False-False-True-True-False] 75.3510μs 33.3787μs 29.9592 KOps/s 29.8975 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[False-False-True-False-True] 67.9210μs 30.5176μs 32.7679 KOps/s 32.2521 KOps/s $\color{#35bf28}+1.60\%$
test_step_mdp_speed[False-False-True-False-False] 52.7510μs 19.7785μs 50.5599 KOps/s 51.3675 KOps/s $\color{#d91a1a}-1.57\%$
test_step_mdp_speed[False-False-False-True-True] 0.1028ms 52.1107μs 19.1899 KOps/s 19.4313 KOps/s $\color{#d91a1a}-1.24\%$
test_step_mdp_speed[False-False-False-True-False] 90.4310μs 35.3425μs 28.2946 KOps/s 28.2759 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[False-False-False-False-True] 66.5410μs 32.9273μs 30.3699 KOps/s 30.5140 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[False-False-False-False-False] 88.5910μs 21.1755μs 47.2244 KOps/s 45.5968 KOps/s $\color{#35bf28}+3.57\%$
test_values[generalized_advantage_estimate-True-True] 26.0496ms 25.6173ms 39.0362 Ops/s 39.2128 Ops/s $\color{#d91a1a}-0.45\%$
test_values[vec_generalized_advantage_estimate-True-True] 97.9165ms 2.8600ms 349.6508 Ops/s 313.0350 Ops/s $\textbf{\color{#35bf28}+11.70\%}$
test_values[td0_return_estimate-False-False] 0.1054ms 80.5601μs 12.4131 KOps/s 12.5408 KOps/s $\color{#d91a1a}-1.02\%$
test_values[td1_return_estimate-False-False] 60.0585ms 57.3153ms 17.4474 Ops/s 17.6770 Ops/s $\color{#d91a1a}-1.30\%$
test_values[vec_td1_return_estimate-False-False] 1.4822ms 1.0983ms 910.5219 Ops/s 914.5041 Ops/s $\color{#d91a1a}-0.44\%$
test_values[td_lambda_return_estimate-True-False] 90.6830ms 90.1236ms 11.0959 Ops/s 11.1303 Ops/s $\color{#d91a1a}-0.31\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2664ms 1.0855ms 921.2713 Ops/s 919.4622 Ops/s $\color{#35bf28}+0.20\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.8354ms 25.4972ms 39.2200 Ops/s 39.9310 Ops/s $\color{#d91a1a}-1.78\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0271ms 0.7559ms 1.3230 KOps/s 1.3145 KOps/s $\color{#35bf28}+0.65\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7616ms 0.6768ms 1.4775 KOps/s 1.4735 KOps/s $\color{#35bf28}+0.27\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5426ms 1.4886ms 671.7794 Ops/s 671.0356 Ops/s $\color{#35bf28}+0.11\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.1010ms 0.6935ms 1.4419 KOps/s 1.4419 KOps/s $-0.00\%$
test_dqn_speed[False-None] 6.9948ms 1.5438ms 647.7425 Ops/s 643.7235 Ops/s $\color{#35bf28}+0.62\%$
test_dqn_speed[False-backward] 2.4913ms 2.1662ms 461.6337 Ops/s 462.6150 Ops/s $\color{#d91a1a}-0.21\%$
test_dqn_speed[True-None] 0.8609ms 0.5763ms 1.7352 KOps/s 1.6886 KOps/s $\color{#35bf28}+2.76\%$
test_dqn_speed[True-backward] 1.2450ms 1.1664ms 857.3167 Ops/s 866.4451 Ops/s $\color{#d91a1a}-1.05\%$
test_dqn_speed[reduce-overhead-None] 0.6847ms 0.6063ms 1.6494 KOps/s 1.6157 KOps/s $\color{#35bf28}+2.09\%$
test_dqn_speed[reduce-overhead-backward] 1.0352ms 0.9818ms 1.0186 KOps/s 1.0013 KOps/s $\color{#35bf28}+1.73\%$
test_ddpg_speed[False-None] 3.2227ms 2.9008ms 344.7353 Ops/s 342.2564 Ops/s $\color{#35bf28}+0.72\%$
test_ddpg_speed[False-backward] 4.7328ms 4.1860ms 238.8907 Ops/s 238.4005 Ops/s $\color{#35bf28}+0.21\%$
test_ddpg_speed[True-None] 1.4647ms 1.3779ms 725.7389 Ops/s 720.8063 Ops/s $\color{#35bf28}+0.68\%$
test_ddpg_speed[True-backward] 2.5357ms 2.4708ms 404.7257 Ops/s 401.6518 Ops/s $\color{#35bf28}+0.77\%$
test_ddpg_speed[reduce-overhead-None] 1.4663ms 1.3836ms 722.7597 Ops/s 717.2978 Ops/s $\color{#35bf28}+0.76\%$
test_ddpg_speed[reduce-overhead-backward] 2.0222ms 1.9248ms 519.5221 Ops/s 506.8617 Ops/s $\color{#35bf28}+2.50\%$
test_sac_speed[False-None] 8.4387ms 8.0687ms 123.9354 Ops/s 121.6569 Ops/s $\color{#35bf28}+1.87\%$
test_sac_speed[False-backward] 11.5082ms 11.0230ms 90.7191 Ops/s 89.1871 Ops/s $\color{#35bf28}+1.72\%$
test_sac_speed[True-None] 2.0602ms 1.8909ms 528.8489 Ops/s 521.5958 Ops/s $\color{#35bf28}+1.39\%$
test_sac_speed[True-backward] 3.7416ms 3.6475ms 274.1631 Ops/s 270.7447 Ops/s $\color{#35bf28}+1.26\%$
test_sac_speed[reduce-overhead-None] 18.1585ms 11.0254ms 90.7000 Ops/s 88.2970 Ops/s $\color{#35bf28}+2.72\%$
test_sac_speed[reduce-overhead-backward] 1.8968ms 1.6806ms 595.0417 Ops/s 533.2744 Ops/s $\textbf{\color{#35bf28}+11.58\%}$
test_redq_speed[False-None] 8.0139ms 7.5927ms 131.7053 Ops/s 129.1950 Ops/s $\color{#35bf28}+1.94\%$
test_redq_speed[False-backward] 12.0323ms 11.3922ms 87.7791 Ops/s 82.7817 Ops/s $\textbf{\color{#35bf28}+6.04\%}$
test_redq_speed[True-None] 2.5562ms 2.3815ms 419.9015 Ops/s 413.4541 Ops/s $\color{#35bf28}+1.56\%$
test_redq_speed[True-backward] 4.3201ms 4.1136ms 243.0961 Ops/s 239.2454 Ops/s $\color{#35bf28}+1.61\%$
test_redq_speed[reduce-overhead-None] 2.5126ms 2.4039ms 415.9880 Ops/s 408.2665 Ops/s $\color{#35bf28}+1.89\%$
test_redq_speed[reduce-overhead-backward] 4.6613ms 4.1456ms 241.2220 Ops/s 227.1179 Ops/s $\textbf{\color{#35bf28}+6.21\%}$
test_redq_deprec_speed[False-None] 9.6768ms 9.2632ms 107.9545 Ops/s 108.3815 Ops/s $\color{#d91a1a}-0.39\%$
test_redq_deprec_speed[False-backward] 12.6896ms 12.1688ms 82.1774 Ops/s 80.3764 Ops/s $\color{#35bf28}+2.24\%$
test_redq_deprec_speed[True-None] 2.9400ms 2.7038ms 369.8508 Ops/s 364.6196 Ops/s $\color{#35bf28}+1.43\%$
test_redq_deprec_speed[True-backward] 5.0581ms 4.6087ms 216.9811 Ops/s 221.6534 Ops/s $\color{#d91a1a}-2.11\%$
test_redq_deprec_speed[reduce-overhead-None] 2.8788ms 2.7068ms 369.4444 Ops/s 364.3913 Ops/s $\color{#35bf28}+1.39\%$
test_redq_deprec_speed[reduce-overhead-backward] 5.0188ms 4.5897ms 217.8800 Ops/s 220.7768 Ops/s $\color{#d91a1a}-1.31\%$
test_td3_speed[False-None] 8.0835ms 7.9883ms 125.1837 Ops/s 123.6329 Ops/s $\color{#35bf28}+1.25\%$
test_td3_speed[False-backward] 11.3224ms 10.5593ms 94.7031 Ops/s 96.0222 Ops/s $\color{#d91a1a}-1.37\%$
test_td3_speed[True-None] 1.8401ms 1.7447ms 573.1747 Ops/s 573.0855 Ops/s $\color{#35bf28}+0.02\%$
test_td3_speed[True-backward] 3.4853ms 3.4313ms 291.4314 Ops/s 285.3639 Ops/s $\color{#35bf28}+2.13\%$
test_td3_speed[reduce-overhead-None] 72.0660ms 28.1424ms 35.5336 Ops/s 35.4900 Ops/s $\color{#35bf28}+0.12\%$
test_td3_speed[reduce-overhead-backward] 1.5896ms 1.5498ms 645.2376 Ops/s 636.1732 Ops/s $\color{#35bf28}+1.42\%$
test_cql_speed[False-None] 17.9248ms 16.9891ms 58.8614 Ops/s 58.4728 Ops/s $\color{#35bf28}+0.66\%$
test_cql_speed[False-backward] 23.5398ms 22.4866ms 44.4710 Ops/s 44.3113 Ops/s $\color{#35bf28}+0.36\%$
test_cql_speed[True-None] 3.8009ms 3.5509ms 281.6219 Ops/s 295.0482 Ops/s $\color{#d91a1a}-4.55\%$
test_cql_speed[True-backward] 6.0758ms 5.7693ms 173.3306 Ops/s 171.1318 Ops/s $\color{#35bf28}+1.28\%$
test_cql_speed[reduce-overhead-None] 19.3311ms 13.2787ms 75.3086 Ops/s 73.0890 Ops/s $\color{#35bf28}+3.04\%$
test_cql_speed[reduce-overhead-backward] 2.2115ms 2.0463ms 488.6838 Ops/s 479.8975 Ops/s $\color{#35bf28}+1.83\%$
test_a2c_speed[False-None] 3.3481ms 3.2180ms 310.7519 Ops/s 303.6799 Ops/s $\color{#35bf28}+2.33\%$
test_a2c_speed[False-backward] 7.4184ms 6.3975ms 156.3121 Ops/s 152.6646 Ops/s $\color{#35bf28}+2.39\%$
test_a2c_speed[True-None] 1.4322ms 1.3762ms 726.6503 Ops/s 715.3457 Ops/s $\color{#35bf28}+1.58\%$
test_a2c_speed[True-backward] 3.1400ms 3.0828ms 324.3752 Ops/s 315.1093 Ops/s $\color{#35bf28}+2.94\%$
test_a2c_speed[reduce-overhead-None] 18.3279ms 8.6843ms 115.1500 Ops/s 114.5487 Ops/s $\color{#35bf28}+0.52\%$
test_a2c_speed[reduce-overhead-backward] 1.7551ms 1.6338ms 612.0651 Ops/s 620.7339 Ops/s $\color{#d91a1a}-1.40\%$
test_ppo_speed[False-None] 3.9426ms 3.7424ms 267.2057 Ops/s 262.7924 Ops/s $\color{#35bf28}+1.68\%$
test_ppo_speed[False-backward] 7.5212ms 7.1012ms 140.8218 Ops/s 136.0102 Ops/s $\color{#35bf28}+3.54\%$
test_ppo_speed[True-None] 1.6566ms 1.4713ms 679.6820 Ops/s 680.5492 Ops/s $\color{#d91a1a}-0.13\%$
test_ppo_speed[True-backward] 3.3666ms 3.2819ms 304.7022 Ops/s 299.8207 Ops/s $\color{#35bf28}+1.63\%$
test_ppo_speed[reduce-overhead-None] 1.3844ms 0.9992ms 1.0009 KOps/s 1.0028 KOps/s $\color{#d91a1a}-0.19\%$
test_ppo_speed[reduce-overhead-backward] 1.7297ms 1.5952ms 626.8938 Ops/s 665.8600 Ops/s $\textbf{\color{#d91a1a}-5.85\%}$
test_reinforce_speed[False-None] 2.6835ms 2.3153ms 431.9089 Ops/s 418.5886 Ops/s $\color{#35bf28}+3.18\%$
test_reinforce_speed[False-backward] 3.8538ms 3.4603ms 288.9896 Ops/s 295.7601 Ops/s $\color{#d91a1a}-2.29\%$
test_reinforce_speed[True-None] 1.7215ms 1.3180ms 758.7147 Ops/s 745.1536 Ops/s $\color{#35bf28}+1.82\%$
test_reinforce_speed[True-backward] 3.1931ms 3.1019ms 322.3824 Ops/s 320.6480 Ops/s $\color{#35bf28}+0.54\%$
test_reinforce_speed[reduce-overhead-None] 16.5261ms 9.4509ms 105.8097 Ops/s 104.8013 Ops/s $\color{#35bf28}+0.96\%$
test_reinforce_speed[reduce-overhead-backward] 1.8170ms 1.6734ms 597.5821 Ops/s 589.8269 Ops/s $\color{#35bf28}+1.31\%$
test_iql_speed[False-None] 9.7365ms 9.2722ms 107.8493 Ops/s 105.7504 Ops/s $\color{#35bf28}+1.98\%$
test_iql_speed[False-backward] 13.7568ms 13.1939ms 75.7924 Ops/s 73.8477 Ops/s $\color{#35bf28}+2.63\%$
test_iql_speed[True-None] 2.6597ms 2.2760ms 439.3593 Ops/s 423.7511 Ops/s $\color{#35bf28}+3.68\%$
test_iql_speed[True-backward] 5.4546ms 5.0142ms 199.4340 Ops/s 193.5254 Ops/s $\color{#35bf28}+3.05\%$
test_iql_speed[reduce-overhead-None] 0.4742s 12.9158ms 77.4243 Ops/s 92.4141 Ops/s $\textbf{\color{#d91a1a}-16.22\%}$
test_iql_speed[reduce-overhead-backward] 2.2144ms 2.1216ms 471.3478 Ops/s 492.0124 Ops/s $\color{#d91a1a}-4.20\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9327ms 6.4090ms 156.0305 Ops/s 154.1309 Ops/s $\color{#35bf28}+1.23\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.4922ms 0.3008ms 3.3246 KOps/s 3.0062 KOps/s $\textbf{\color{#35bf28}+10.59\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4750ms 0.2620ms 3.8172 KOps/s 3.2068 KOps/s $\textbf{\color{#35bf28}+19.04\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.7049ms 6.1501ms 162.6001 Ops/s 161.4951 Ops/s $\color{#35bf28}+0.68\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.1295ms 0.3182ms 3.1424 KOps/s 3.6984 KOps/s $\textbf{\color{#d91a1a}-15.03\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5255ms 0.2840ms 3.5210 KOps/s 4.0088 KOps/s $\textbf{\color{#d91a1a}-12.17\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4765ms 1.2679ms 788.7343 Ops/s 789.3977 Ops/s $\color{#d91a1a}-0.08\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4967ms 1.1664ms 857.3440 Ops/s 844.0383 Ops/s $\color{#35bf28}+1.58\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5850ms 6.3278ms 158.0318 Ops/s 157.5429 Ops/s $\color{#35bf28}+0.31\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.9807ms 0.4337ms 2.3057 KOps/s 2.3115 KOps/s $\color{#d91a1a}-0.25\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7062ms 0.4246ms 2.3552 KOps/s 2.1989 KOps/s $\textbf{\color{#35bf28}+7.11\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 10.0666ms 6.3478ms 157.5349 Ops/s 161.2558 Ops/s $\color{#d91a1a}-2.31\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.2903ms 0.3642ms 2.7456 KOps/s 2.9561 KOps/s $\textbf{\color{#d91a1a}-7.12\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9120ms 0.3266ms 3.0618 KOps/s 2.9748 KOps/s $\color{#35bf28}+2.92\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3919ms 6.0816ms 164.4314 Ops/s 162.3017 Ops/s $\color{#35bf28}+1.31\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0555ms 0.3495ms 2.8609 KOps/s 3.5620 KOps/s $\textbf{\color{#d91a1a}-19.68\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4952ms 0.3012ms 3.3203 KOps/s 4.0313 KOps/s $\textbf{\color{#d91a1a}-17.64\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.5013ms 6.2885ms 159.0204 Ops/s 157.6365 Ops/s $\color{#35bf28}+0.88\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1280ms 0.3995ms 2.5029 KOps/s 2.2181 KOps/s $\textbf{\color{#35bf28}+12.84\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6464ms 0.4568ms 2.1894 KOps/s 2.5749 KOps/s $\textbf{\color{#d91a1a}-14.97\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0190ms 5.4404ms 183.8111 Ops/s 180.1563 Ops/s $\color{#35bf28}+2.03\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.0748ms 1.9309ms 517.8984 Ops/s 427.8155 Ops/s $\textbf{\color{#35bf28}+21.06\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.8746ms 1.2501ms 799.9135 Ops/s 845.0812 Ops/s $\textbf{\color{#d91a1a}-5.34\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4528s 14.8388ms 67.3910 Ops/s 179.9797 Ops/s $\textbf{\color{#d91a1a}-62.56\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.9513ms 1.9460ms 513.8833 Ops/s 430.6541 Ops/s $\textbf{\color{#35bf28}+19.33\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.7070ms 1.2172ms 821.5885 Ops/s 829.0372 Ops/s $\color{#d91a1a}-0.90\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.7538ms 5.6962ms 175.5566 Ops/s 30.8762 Ops/s $\textbf{\color{#35bf28}+468.58\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.9805ms 2.2040ms 453.7263 Ops/s 432.4881 Ops/s $\color{#35bf28}+4.91\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.3557ms 1.3945ms 717.1264 Ops/s 713.7360 Ops/s $\color{#35bf28}+0.48\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.3889ms 13.1124ms 76.2638 Ops/s 72.1069 Ops/s $\textbf{\color{#35bf28}+5.76\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.5065ms 16.7599ms 59.6662 Ops/s 58.9044 Ops/s $\color{#35bf28}+1.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.1268ms 17.7649ms 56.2906 Ops/s 54.2156 Ops/s $\color{#35bf28}+3.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.3054ms 16.8944ms 59.1912 Ops/s 58.4642 Ops/s $\color{#35bf28}+1.24\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.3151ms 17.7821ms 56.2364 Ops/s 54.7006 Ops/s $\color{#35bf28}+2.81\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.8481ms 18.5479ms 53.9144 Ops/s 53.5192 Ops/s $\color{#35bf28}+0.74\%$

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: a3b2a95cc95a302f5c0dba4d16f8fe2b3fecf518
Pull Request resolved: #2788
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: 1df9caf353241f501412f4cfd73ac10e0f9f97a5
Pull Request resolved: #2788
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: f3e82fed162415a131170fa50c2d467bf6bcafff
Pull Request resolved: #2788
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: dc00ea3d23e015756974cd5c2ce638b55e5f6f92
Pull Request resolved: #2788
@vmoens vmoens merged commit 3730096 into gh/vmoens/91/base Feb 13, 2025
71 of 74 checks passed
vmoens added a commit that referenced this pull request Feb 13, 2025
ghstack-source-id: dc00ea3d23e015756974cd5c2ce638b55e5f6f92
Pull Request resolved: #2788
@vmoens vmoens deleted the gh/vmoens/91/head branch February 13, 2025 17:44
@vmoens vmoens added the enhancement New feature or request label Feb 13, 2025
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. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants