Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Jan 23, 2026

Add a label based on the [] prefix in title

@pytorch-bot
Copy link

pytorch-bot bot commented Jan 23, 2026

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3381

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

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

@meta-cla meta-cla 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 Jan 23, 2026
@github-actions github-actions bot added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Jan 23, 2026
@vmoens vmoens merged commit da87455 into main Jan 23, 2026
28 of 36 checks passed
@vmoens vmoens deleted the autotag branch January 23, 2026 09:24
@github-actions
Copy link

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

Total Benchmarks: 148. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}22$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.9990μs 83.3445μs 11.9984 KOps/s 12.1154 KOps/s $\color{#d91a1a}-0.97\%$
test_tensor_to_bytestream_speed[torch.save] 0.1396ms 0.1388ms 7.2068 KOps/s 7.3663 KOps/s $\color{#d91a1a}-2.17\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1040s 0.1030s 9.7099 Ops/s 9.2567 Ops/s $\color{#35bf28}+4.90\%$
test_tensor_to_bytestream_speed[numpy] 2.3857μs 2.3800μs 420.1594 KOps/s 422.5213 KOps/s $\color{#d91a1a}-0.56\%$
test_tensor_to_bytestream_speed[safetensors] 38.1311μs 37.8054μs 26.4512 KOps/s 27.6555 KOps/s $\color{#d91a1a}-4.35\%$
test_simple 0.8903s 0.7968s 1.2550 Ops/s 1.2971 Ops/s $\color{#d91a1a}-3.25\%$
test_transformed 1.5030s 1.4125s 0.7080 Ops/s 0.7229 Ops/s $\color{#d91a1a}-2.06\%$
test_serial 2.3521s 2.2963s 0.4355 Ops/s 0.4490 Ops/s $\color{#d91a1a}-3.00\%$
test_parallel 2.0148s 1.9202s 0.5208 Ops/s 0.5227 Ops/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[True-True-True-True-True] 0.2776ms 45.6343μs 21.9133 KOps/s 22.9731 KOps/s $\color{#d91a1a}-4.61\%$
test_step_mdp_speed[True-True-True-True-False] 92.5310μs 25.4375μs 39.3120 KOps/s 40.9351 KOps/s $\color{#d91a1a}-3.97\%$
test_step_mdp_speed[True-True-True-False-True] 95.5420μs 24.6066μs 40.6395 KOps/s 41.3800 KOps/s $\color{#d91a1a}-1.79\%$
test_step_mdp_speed[True-True-True-False-False] 41.7210μs 14.0513μs 71.1677 KOps/s 74.6171 KOps/s $\color{#d91a1a}-4.62\%$
test_step_mdp_speed[True-True-False-True-True] 87.4910μs 48.4492μs 20.6402 KOps/s 21.6805 KOps/s $\color{#d91a1a}-4.80\%$
test_step_mdp_speed[True-True-False-True-False] 58.6810μs 28.3116μs 35.3212 KOps/s 36.2162 KOps/s $\color{#d91a1a}-2.47\%$
test_step_mdp_speed[True-True-False-False-True] 61.3610μs 27.7751μs 36.0034 KOps/s 36.6974 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[True-True-False-False-False] 43.3700μs 16.8587μs 59.3165 KOps/s 61.6602 KOps/s $\color{#d91a1a}-3.80\%$
test_step_mdp_speed[True-False-True-True-True] 89.5010μs 51.2985μs 19.4937 KOps/s 20.2443 KOps/s $\color{#d91a1a}-3.71\%$
test_step_mdp_speed[True-False-True-True-False] 60.0510μs 31.0942μs 32.1603 KOps/s 33.1987 KOps/s $\color{#d91a1a}-3.13\%$
test_step_mdp_speed[True-False-True-False-True] 57.9710μs 28.2670μs 35.3770 KOps/s 37.1366 KOps/s $\color{#d91a1a}-4.74\%$
test_step_mdp_speed[True-False-True-False-False] 55.5010μs 16.8226μs 59.4440 KOps/s 63.1770 KOps/s $\textbf{\color{#d91a1a}-5.91\%}$
test_step_mdp_speed[True-False-False-True-True] 94.3520μs 54.1075μs 18.4817 KOps/s 19.2446 KOps/s $\color{#d91a1a}-3.96\%$
test_step_mdp_speed[True-False-False-True-False] 76.2910μs 33.0671μs 30.2416 KOps/s 30.9369 KOps/s $\color{#d91a1a}-2.25\%$
test_step_mdp_speed[True-False-False-False-True] 62.5510μs 31.3296μs 31.9187 KOps/s 34.2121 KOps/s $\textbf{\color{#d91a1a}-6.70\%}$
test_step_mdp_speed[True-False-False-False-False] 47.7700μs 19.6289μs 50.9452 KOps/s 54.3396 KOps/s $\textbf{\color{#d91a1a}-6.25\%}$
test_step_mdp_speed[False-True-True-True-True] 93.7920μs 51.3762μs 19.4643 KOps/s 20.7256 KOps/s $\textbf{\color{#d91a1a}-6.09\%}$
test_step_mdp_speed[False-True-True-True-False] 68.6510μs 30.6702μs 32.6049 KOps/s 34.1344 KOps/s $\color{#d91a1a}-4.48\%$
test_step_mdp_speed[False-True-True-False-True] 75.0010μs 32.4091μs 30.8555 KOps/s 33.4530 KOps/s $\textbf{\color{#d91a1a}-7.76\%}$
test_step_mdp_speed[False-True-True-False-False] 67.6810μs 18.3410μs 54.5225 KOps/s 56.7140 KOps/s $\color{#d91a1a}-3.86\%$
test_step_mdp_speed[False-True-False-True-True] 2.5991ms 55.3091μs 18.0802 KOps/s 19.4618 KOps/s $\textbf{\color{#d91a1a}-7.10\%}$
test_step_mdp_speed[False-True-False-True-False] 73.2510μs 34.3693μs 29.0958 KOps/s 30.9680 KOps/s $\textbf{\color{#d91a1a}-6.05\%}$
test_step_mdp_speed[False-True-False-False-True] 70.0410μs 34.4009μs 29.0690 KOps/s 30.0360 KOps/s $\color{#d91a1a}-3.22\%$
test_step_mdp_speed[False-True-False-False-False] 50.4210μs 21.4708μs 46.5750 KOps/s 49.8287 KOps/s $\textbf{\color{#d91a1a}-6.53\%}$
test_step_mdp_speed[False-False-True-True-True] 86.6210μs 56.7403μs 17.6242 KOps/s 18.3378 KOps/s $\color{#d91a1a}-3.89\%$
test_step_mdp_speed[False-False-True-True-False] 72.4310μs 36.3081μs 27.5421 KOps/s 28.6207 KOps/s $\color{#d91a1a}-3.77\%$
test_step_mdp_speed[False-False-True-False-True] 71.5610μs 35.4840μs 28.1817 KOps/s 30.4415 KOps/s $\textbf{\color{#d91a1a}-7.42\%}$
test_step_mdp_speed[False-False-True-False-False] 51.7300μs 21.8167μs 45.8364 KOps/s 48.7449 KOps/s $\textbf{\color{#d91a1a}-5.97\%}$
test_step_mdp_speed[False-False-False-True-True] 95.4710μs 60.2303μs 16.6029 KOps/s 17.7479 KOps/s $\textbf{\color{#d91a1a}-6.45\%}$
test_step_mdp_speed[False-False-False-True-False] 72.9610μs 38.9912μs 25.6468 KOps/s 26.3380 KOps/s $\color{#d91a1a}-2.62\%$
test_step_mdp_speed[False-False-False-False-True] 76.4610μs 37.3410μs 26.7802 KOps/s 28.4187 KOps/s $\textbf{\color{#d91a1a}-5.77\%}$
test_step_mdp_speed[False-False-False-False-False] 64.7510μs 24.3874μs 41.0048 KOps/s 43.4164 KOps/s $\textbf{\color{#d91a1a}-5.55\%}$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8486s 0.7554s 1.3239 Ops/s 1.3318 Ops/s $\color{#d91a1a}-0.60\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7205s 0.6228s 1.6057 Ops/s 1.6064 Ops/s $\color{#d91a1a}-0.04\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7141s 1.6394s 0.6100 Ops/s 0.6100 Ops/s $-0.00\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.4990s 1.4236s 0.7025 Ops/s 0.7031 Ops/s $\color{#d91a1a}-0.09\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9645s 1.8829s 0.5311 Ops/s 0.5328 Ops/s $\color{#d91a1a}-0.32\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7400s 1.6631s 0.6013 Ops/s 0.5997 Ops/s $\color{#35bf28}+0.26\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.6938s 4.5051s 0.2220 Ops/s 0.2202 Ops/s $\color{#35bf28}+0.79\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4639s 4.3816s 0.2282 Ops/s 0.2275 Ops/s $\color{#35bf28}+0.34\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0040s 1.9236s 0.5199 Ops/s 0.5232 Ops/s $\color{#d91a1a}-0.63\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7004s 1.6212s 0.6168 Ops/s 0.6146 Ops/s $\color{#35bf28}+0.36\%$
test_values[generalized_advantage_estimate-True-True] 19.6720ms 19.2147ms 52.0435 Ops/s 48.0045 Ops/s $\textbf{\color{#35bf28}+8.41\%}$
test_values[vec_generalized_advantage_estimate-True-True] 0.1418s 3.7309ms 268.0304 Ops/s 280.5233 Ops/s $\color{#d91a1a}-4.45\%$
test_values[td0_return_estimate-False-False] 0.1007ms 79.1905μs 12.6278 KOps/s 12.5254 KOps/s $\color{#35bf28}+0.82\%$
test_values[td1_return_estimate-False-False] 45.5630ms 45.2654ms 22.0919 Ops/s 21.5097 Ops/s $\color{#35bf28}+2.71\%$
test_values[vec_td1_return_estimate-False-False] 1.2961ms 1.0572ms 945.8989 Ops/s 938.8762 Ops/s $\color{#35bf28}+0.75\%$
test_values[td_lambda_return_estimate-True-False] 74.9962ms 74.5803ms 13.4084 Ops/s 13.1257 Ops/s $\color{#35bf28}+2.15\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3139ms 1.0556ms 947.2851 Ops/s 943.7764 Ops/s $\color{#35bf28}+0.37\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 19.6449ms 19.4216ms 51.4890 Ops/s 50.5552 Ops/s $\color{#35bf28}+1.85\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0017ms 0.7281ms 1.3734 KOps/s 1.3653 KOps/s $\color{#35bf28}+0.59\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7018ms 0.6509ms 1.5362 KOps/s 1.5217 KOps/s $\color{#35bf28}+0.95\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5627ms 1.4659ms 682.1623 Ops/s 679.7251 Ops/s $\color{#35bf28}+0.36\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7155ms 0.6673ms 1.4986 KOps/s 1.4756 KOps/s $\color{#35bf28}+1.56\%$
test_dqn_speed[False-None] 1.6396ms 1.5015ms 665.9831 Ops/s 666.4378 Ops/s $\color{#d91a1a}-0.07\%$
test_dqn_speed[False-backward] 2.1865ms 2.1115ms 473.5916 Ops/s 473.3417 Ops/s $\color{#35bf28}+0.05\%$
test_dqn_speed[True-None] 0.6376ms 0.5677ms 1.7614 KOps/s 1.7593 KOps/s $\color{#35bf28}+0.12\%$
test_dqn_speed[True-backward] 1.2533ms 1.1859ms 843.2505 Ops/s 843.4500 Ops/s $\color{#d91a1a}-0.02\%$
test_dqn_speed[reduce-overhead-None] 0.6602ms 0.5884ms 1.6996 KOps/s 1.6651 KOps/s $\color{#35bf28}+2.07\%$
test_ddpg_speed[False-None] 3.2180ms 2.8178ms 354.8923 Ops/s 351.0275 Ops/s $\color{#35bf28}+1.10\%$
test_ddpg_speed[False-backward] 4.5333ms 4.1659ms 240.0468 Ops/s 238.3254 Ops/s $\color{#35bf28}+0.72\%$
test_ddpg_speed[True-None] 1.3359ms 1.2905ms 774.9072 Ops/s 761.2606 Ops/s $\color{#35bf28}+1.79\%$
test_ddpg_speed[True-backward] 2.8235ms 2.5105ms 398.3315 Ops/s 395.7304 Ops/s $\color{#35bf28}+0.66\%$
test_ddpg_speed[reduce-overhead-None] 1.5777ms 1.3721ms 728.8281 Ops/s 742.7956 Ops/s $\color{#d91a1a}-1.88\%$
test_sac_speed[False-None] 8.7513ms 8.3909ms 119.1769 Ops/s 122.5974 Ops/s $\color{#d91a1a}-2.79\%$
test_sac_speed[False-backward] 12.0855ms 11.3706ms 87.9464 Ops/s 88.2272 Ops/s $\color{#d91a1a}-0.32\%$
test_sac_speed[True-None] 1.8857ms 1.8005ms 555.4160 Ops/s 551.5650 Ops/s $\color{#35bf28}+0.70\%$
test_sac_speed[True-backward] 3.6777ms 3.5739ms 279.8096 Ops/s 277.5110 Ops/s $\color{#35bf28}+0.83\%$
test_sac_speed[reduce-overhead-None] 0.3891s 11.5620ms 86.4904 Ops/s 95.9455 Ops/s $\textbf{\color{#d91a1a}-9.85\%}$
test_redq_deprec_speed[False-None] 9.7144ms 9.0913ms 109.9955 Ops/s 108.9465 Ops/s $\color{#35bf28}+0.96\%$
test_redq_deprec_speed[False-backward] 12.8840ms 12.4239ms 80.4898 Ops/s 80.6231 Ops/s $\color{#d91a1a}-0.17\%$
test_redq_deprec_speed[True-None] 2.6940ms 2.4995ms 400.0776 Ops/s 385.7502 Ops/s $\color{#35bf28}+3.71\%$
test_redq_deprec_speed[True-backward] 4.6263ms 4.2414ms 235.7690 Ops/s 232.6601 Ops/s $\color{#35bf28}+1.34\%$
test_redq_deprec_speed[reduce-overhead-None] 0.4185s 10.9206ms 91.5697 Ops/s 124.3760 Ops/s $\textbf{\color{#d91a1a}-26.38\%}$
test_td3_speed[False-None] 8.0991ms 7.9648ms 125.5532 Ops/s 122.5910 Ops/s $\color{#35bf28}+2.42\%$
test_td3_speed[False-backward] 11.2402ms 10.5552ms 94.7397 Ops/s 93.9307 Ops/s $\color{#35bf28}+0.86\%$
test_td3_speed[True-None] 1.6618ms 1.6420ms 609.0062 Ops/s 609.5136 Ops/s $\color{#d91a1a}-0.08\%$
test_td3_speed[True-backward] 3.3987ms 3.2895ms 303.9988 Ops/s 321.2211 Ops/s $\textbf{\color{#d91a1a}-5.36\%}$
test_td3_speed[reduce-overhead-None] 82.8581ms 23.2501ms 43.0105 Ops/s 43.1250 Ops/s $\color{#d91a1a}-0.27\%$
test_cql_speed[False-None] 18.5043ms 16.9404ms 59.0306 Ops/s 59.2315 Ops/s $\color{#d91a1a}-0.34\%$
test_cql_speed[False-backward] 23.2468ms 22.3130ms 44.8169 Ops/s 45.6993 Ops/s $\color{#d91a1a}-1.93\%$
test_cql_speed[True-None] 3.7724ms 3.3787ms 295.9701 Ops/s 300.7707 Ops/s $\color{#d91a1a}-1.60\%$
test_cql_speed[True-backward] 6.0878ms 5.5480ms 180.2448 Ops/s 179.0945 Ops/s $\color{#35bf28}+0.64\%$
test_cql_speed[reduce-overhead-None] 18.5546ms 11.4420ms 87.3971 Ops/s 87.9133 Ops/s $\color{#d91a1a}-0.59\%$
test_a2c_speed[False-None] 3.8624ms 3.1524ms 317.2171 Ops/s 315.7334 Ops/s $\color{#35bf28}+0.47\%$
test_a2c_speed[False-backward] 6.7085ms 6.1999ms 161.2933 Ops/s 159.8493 Ops/s $\color{#35bf28}+0.90\%$
test_a2c_speed[True-None] 1.5309ms 1.3152ms 760.3642 Ops/s 747.0194 Ops/s $\color{#35bf28}+1.79\%$
test_a2c_speed[True-backward] 3.2547ms 3.1114ms 321.3972 Ops/s 321.7636 Ops/s $\color{#d91a1a}-0.11\%$
test_a2c_speed[reduce-overhead-None] 1.0790ms 0.9471ms 1.0558 KOps/s 1.0468 KOps/s $\color{#35bf28}+0.86\%$
test_ppo_speed[False-None] 3.9488ms 3.7526ms 266.4852 Ops/s 267.5591 Ops/s $\color{#d91a1a}-0.40\%$
test_ppo_speed[False-backward] 7.4002ms 6.9813ms 143.2392 Ops/s 141.9867 Ops/s $\color{#35bf28}+0.88\%$
test_ppo_speed[True-None] 1.5460ms 1.3944ms 717.1481 Ops/s 701.9299 Ops/s $\color{#35bf28}+2.17\%$
test_ppo_speed[True-backward] 3.1053ms 3.0521ms 327.6404 Ops/s 304.3793 Ops/s $\textbf{\color{#35bf28}+7.64\%}$
test_ppo_speed[reduce-overhead-None] 1.2035ms 1.0058ms 994.1977 Ops/s 962.3859 Ops/s $\color{#35bf28}+3.31\%$
test_reinforce_speed[False-None] 2.5215ms 2.2483ms 444.7862 Ops/s 446.2246 Ops/s $\color{#d91a1a}-0.32\%$
test_reinforce_speed[False-backward] 3.8834ms 3.4311ms 291.4512 Ops/s 310.1131 Ops/s $\textbf{\color{#d91a1a}-6.02\%}$
test_reinforce_speed[True-None] 1.7217ms 1.2342ms 810.2306 Ops/s 770.5673 Ops/s $\textbf{\color{#35bf28}+5.15\%}$
test_reinforce_speed[True-backward] 3.1230ms 3.0476ms 328.1226 Ops/s 340.5795 Ops/s $\color{#d91a1a}-3.66\%$
test_reinforce_speed[reduce-overhead-None] 0.4622s 9.8528ms 101.4936 Ops/s 97.4551 Ops/s $\color{#35bf28}+4.14\%$
test_iql_speed[False-None] 9.7863ms 9.1405ms 109.4033 Ops/s 108.2662 Ops/s $\color{#35bf28}+1.05\%$
test_iql_speed[False-backward] 13.6679ms 13.0632ms 76.5506 Ops/s 77.6725 Ops/s $\color{#d91a1a}-1.44\%$
test_iql_speed[True-None] 3.4270ms 2.1852ms 457.6251 Ops/s 454.4288 Ops/s $\color{#35bf28}+0.70\%$
test_iql_speed[True-backward] 5.2774ms 4.8631ms 205.6315 Ops/s 206.6426 Ops/s $\color{#d91a1a}-0.49\%$
test_iql_speed[reduce-overhead-None] 14.6470ms 8.8544ms 112.9388 Ops/s 77.1602 Ops/s $\textbf{\color{#35bf28}+46.37\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2742ms 5.8257ms 171.6524 Ops/s 171.6336 Ops/s $\color{#35bf28}+0.01\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0443ms 0.3650ms 2.7397 KOps/s 2.8512 KOps/s $\color{#d91a1a}-3.91\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6120ms 0.3348ms 2.9871 KOps/s 2.9982 KOps/s $\color{#d91a1a}-0.37\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0737ms 5.6556ms 176.8157 Ops/s 174.1759 Ops/s $\color{#35bf28}+1.52\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9244ms 0.3067ms 3.2606 KOps/s 3.1178 KOps/s $\color{#35bf28}+4.58\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5805ms 0.3191ms 3.1339 KOps/s 3.1972 KOps/s $\color{#d91a1a}-1.98\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5602ms 1.3628ms 733.7772 Ops/s 732.6787 Ops/s $\color{#35bf28}+0.15\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4895ms 1.2484ms 801.0401 Ops/s 790.1960 Ops/s $\color{#35bf28}+1.37\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.7540ms 5.9694ms 167.5201 Ops/s 169.4950 Ops/s $\color{#d91a1a}-1.17\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8563ms 0.4605ms 2.1715 KOps/s 2.0452 KOps/s $\textbf{\color{#35bf28}+6.18\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6638ms 0.4273ms 2.3403 KOps/s 2.0144 KOps/s $\textbf{\color{#35bf28}+16.17\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8474ms 5.6473ms 177.0760 Ops/s 173.7829 Ops/s $\color{#35bf28}+1.89\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8590ms 0.3677ms 2.7194 KOps/s 2.8817 KOps/s $\textbf{\color{#d91a1a}-5.63\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5537ms 0.3380ms 2.9589 KOps/s 3.0375 KOps/s $\color{#d91a1a}-2.59\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9603ms 5.6153ms 178.0851 Ops/s 175.5505 Ops/s $\color{#35bf28}+1.44\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2803ms 0.3578ms 2.7948 KOps/s 3.6112 KOps/s $\textbf{\color{#d91a1a}-22.61\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5006ms 0.3349ms 2.9857 KOps/s 3.9081 KOps/s $\textbf{\color{#d91a1a}-23.60\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0581ms 5.8211ms 171.7884 Ops/s 169.5041 Ops/s $\color{#35bf28}+1.35\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.3169ms 0.5008ms 1.9969 KOps/s 1.9442 KOps/s $\color{#35bf28}+2.71\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6078ms 0.4328ms 2.3106 KOps/s 2.0462 KOps/s $\textbf{\color{#35bf28}+12.92\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4741ms 4.9873ms 200.5111 Ops/s 197.0259 Ops/s $\color{#35bf28}+1.77\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 4.4714ms 2.1052ms 475.0218 Ops/s 501.4205 Ops/s $\textbf{\color{#d91a1a}-5.26\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.2859ms 0.9205ms 1.0863 KOps/s 1.1086 KOps/s $\color{#d91a1a}-2.01\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5852s 16.7182ms 59.8152 Ops/s 196.1401 Ops/s $\textbf{\color{#d91a1a}-69.50\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.3033ms 1.9786ms 505.4039 Ops/s 514.8581 Ops/s $\color{#d91a1a}-1.84\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 11.0386ms 1.2996ms 769.4646 Ops/s 788.8267 Ops/s $\color{#d91a1a}-2.45\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.8092ms 5.2048ms 192.1301 Ops/s 49.4135 Ops/s $\textbf{\color{#35bf28}+288.82\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.9174ms 2.0375ms 490.7882 Ops/s 465.1405 Ops/s $\textbf{\color{#35bf28}+5.51\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 3.0951ms 1.0873ms 919.7234 Ops/s 931.7519 Ops/s $\color{#d91a1a}-1.29\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.6759ms 35.2725ms 28.3507 Ops/s 28.0606 Ops/s $\color{#35bf28}+1.03\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.4644ms 17.6618ms 56.6195 Ops/s 56.5442 Ops/s $\color{#35bf28}+0.13\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 40.1293ms 36.6058ms 27.3180 Ops/s 27.3196 Ops/s $-0.01\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.6845ms 18.0907ms 55.2769 Ops/s 56.3701 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 40.0686ms 38.1919ms 26.1835 Ops/s 26.2009 Ops/s $\color{#d91a1a}-0.07\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.9314ms 19.5344ms 51.1916 Ops/s 51.2936 Ops/s $\color{#d91a1a}-0.20\%$

@github-actions
Copy link

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

Total Benchmarks: 153. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.0250μs 80.6003μs 12.4069 KOps/s 12.0210 KOps/s $\color{#35bf28}+3.21\%$
test_tensor_to_bytestream_speed[torch.save] 0.1392ms 0.1387ms 7.2083 KOps/s 7.1098 KOps/s $\color{#35bf28}+1.39\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1096s 0.1095s 9.1360 Ops/s 9.3534 Ops/s $\color{#d91a1a}-2.32\%$
test_tensor_to_bytestream_speed[numpy] 2.5209μs 2.5183μs 397.0880 KOps/s 403.7094 KOps/s $\color{#d91a1a}-1.64\%$
test_tensor_to_bytestream_speed[safetensors] 37.9584μs 36.9892μs 27.0349 KOps/s 27.0133 KOps/s $\color{#35bf28}+0.08\%$
test_simple 0.5539s 0.5510s 1.8150 Ops/s 1.8091 Ops/s $\color{#35bf28}+0.33\%$
test_transformed 1.2426s 1.1534s 0.8670 Ops/s 0.8651 Ops/s $\color{#35bf28}+0.22\%$
test_serial 1.6900s 1.6869s 0.5928 Ops/s 0.5945 Ops/s $\color{#d91a1a}-0.29\%$
test_parallel 1.2178s 1.1343s 0.8816 Ops/s 0.8972 Ops/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[True-True-True-True-True] 0.2800ms 45.5089μs 21.9737 KOps/s 22.3776 KOps/s $\color{#d91a1a}-1.80\%$
test_step_mdp_speed[True-True-True-True-False] 57.4400μs 24.8211μs 40.2882 KOps/s 39.4022 KOps/s $\color{#35bf28}+2.25\%$
test_step_mdp_speed[True-True-True-False-True] 89.8010μs 25.5346μs 39.1626 KOps/s 38.3013 KOps/s $\color{#35bf28}+2.25\%$
test_step_mdp_speed[True-True-True-False-False] 67.4610μs 14.0591μs 71.1284 KOps/s 71.8130 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[True-True-False-True-True] 89.2310μs 48.3949μs 20.6634 KOps/s 20.4488 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-True-False-True-False] 59.3310μs 27.5738μs 36.2663 KOps/s 35.5939 KOps/s $\color{#35bf28}+1.89\%$
test_step_mdp_speed[True-True-False-False-True] 75.0310μs 28.4137μs 35.1943 KOps/s 35.2180 KOps/s $\color{#d91a1a}-0.07\%$
test_step_mdp_speed[True-True-False-False-False] 48.2010μs 16.7312μs 59.7687 KOps/s 59.1686 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[True-False-True-True-True] 93.2810μs 52.6554μs 18.9914 KOps/s 19.2545 KOps/s $\color{#d91a1a}-1.37\%$
test_step_mdp_speed[True-False-True-True-False] 63.3710μs 30.4610μs 32.8289 KOps/s 32.2490 KOps/s $\color{#35bf28}+1.80\%$
test_step_mdp_speed[True-False-True-False-True] 61.3610μs 27.6471μs 36.1702 KOps/s 35.6128 KOps/s $\color{#35bf28}+1.56\%$
test_step_mdp_speed[True-False-True-False-False] 47.8800μs 16.5823μs 60.3054 KOps/s 59.5013 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[True-False-False-True-True] 89.6510μs 53.5285μs 18.6816 KOps/s 18.5023 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-False-False-True-False] 69.0510μs 33.0339μs 30.2720 KOps/s 29.9059 KOps/s $\color{#35bf28}+1.22\%$
test_step_mdp_speed[True-False-False-False-True] 96.6420μs 29.6991μs 33.6711 KOps/s 32.3741 KOps/s $\color{#35bf28}+4.01\%$
test_step_mdp_speed[True-False-False-False-False] 89.6610μs 19.1903μs 52.1097 KOps/s 51.1979 KOps/s $\color{#35bf28}+1.78\%$
test_step_mdp_speed[False-True-True-True-True] 83.3910μs 50.4562μs 19.8192 KOps/s 19.6529 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[False-True-True-True-False] 79.7510μs 30.8461μs 32.4190 KOps/s 32.7000 KOps/s $\color{#d91a1a}-0.86\%$
test_step_mdp_speed[False-True-True-False-True] 62.5300μs 31.8330μs 31.4140 KOps/s 30.5310 KOps/s $\color{#35bf28}+2.89\%$
test_step_mdp_speed[False-True-True-False-False] 0.1766ms 18.4526μs 54.1928 KOps/s 53.4679 KOps/s $\color{#35bf28}+1.36\%$
test_step_mdp_speed[False-True-False-True-True] 2.8479ms 52.9036μs 18.9023 KOps/s 18.6263 KOps/s $\color{#35bf28}+1.48\%$
test_step_mdp_speed[False-True-False-True-False] 75.2110μs 32.8188μs 30.4703 KOps/s 30.2553 KOps/s $\color{#35bf28}+0.71\%$
test_step_mdp_speed[False-True-False-False-True] 75.1510μs 33.8151μs 29.5726 KOps/s 29.0400 KOps/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[False-True-False-False-False] 94.2910μs 21.3845μs 46.7628 KOps/s 47.2260 KOps/s $\color{#d91a1a}-0.98\%$
test_step_mdp_speed[False-False-True-True-True] 91.7710μs 55.6587μs 17.9666 KOps/s 18.0865 KOps/s $\color{#d91a1a}-0.66\%$
test_step_mdp_speed[False-False-True-True-False] 64.5310μs 35.8320μs 27.9080 KOps/s 27.7137 KOps/s $\color{#35bf28}+0.70\%$
test_step_mdp_speed[False-False-True-False-True] 93.5610μs 34.3662μs 29.0984 KOps/s 29.1018 KOps/s $\color{#d91a1a}-0.01\%$
test_step_mdp_speed[False-False-True-False-False] 50.9410μs 20.8544μs 47.9514 KOps/s 46.7495 KOps/s $\color{#35bf28}+2.57\%$
test_step_mdp_speed[False-False-False-True-True] 93.7510μs 57.4194μs 17.4157 KOps/s 17.3174 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[False-False-False-True-False] 73.3310μs 38.1336μs 26.2236 KOps/s 25.7952 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[False-False-False-False-True] 81.3510μs 36.8746μs 27.1189 KOps/s 27.5492 KOps/s $\color{#d91a1a}-1.56\%$
test_step_mdp_speed[False-False-False-False-False] 78.0810μs 23.2142μs 43.0772 KOps/s 41.6726 KOps/s $\color{#35bf28}+3.37\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8623s 0.7663s 1.3049 Ops/s 1.3036 Ops/s $\color{#35bf28}+0.10\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7249s 0.6321s 1.5819 Ops/s 1.5720 Ops/s $\color{#35bf28}+0.63\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7418s 1.6701s 0.5988 Ops/s 0.5959 Ops/s $\color{#35bf28}+0.47\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5210s 1.4458s 0.6917 Ops/s 0.6865 Ops/s $\color{#35bf28}+0.75\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9887s 1.9203s 0.5208 Ops/s 0.5187 Ops/s $\color{#35bf28}+0.40\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7794s 1.7045s 0.5867 Ops/s 0.5833 Ops/s $\color{#35bf28}+0.58\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.7896s 4.6302s 0.2160 Ops/s 0.2149 Ops/s $\color{#35bf28}+0.51\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6156s 4.4782s 0.2233 Ops/s 0.2214 Ops/s $\color{#35bf28}+0.86\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0329s 1.9442s 0.5143 Ops/s 0.5119 Ops/s $\color{#35bf28}+0.48\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7648s 1.6771s 0.5963 Ops/s 0.5973 Ops/s $\color{#d91a1a}-0.17\%$
test_values[generalized_advantage_estimate-True-True] 10.7918ms 10.4653ms 95.5540 Ops/s 92.3750 Ops/s $\color{#35bf28}+3.44\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.4787ms 17.5056ms 57.1244 Ops/s 56.9175 Ops/s $\color{#35bf28}+0.36\%$
test_values[td0_return_estimate-False-False] 0.2242ms 0.1320ms 7.5782 KOps/s 7.6449 KOps/s $\color{#d91a1a}-0.87\%$
test_values[td1_return_estimate-False-False] 29.6505ms 28.3442ms 35.2806 Ops/s 34.0226 Ops/s $\color{#35bf28}+3.70\%$
test_values[vec_td1_return_estimate-False-False] 19.1527ms 17.5660ms 56.9281 Ops/s 56.7139 Ops/s $\color{#35bf28}+0.38\%$
test_values[td_lambda_return_estimate-True-False] 43.7781ms 42.1922ms 23.7011 Ops/s 22.6503 Ops/s $\color{#35bf28}+4.64\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.4173ms 17.6082ms 56.7916 Ops/s 56.6677 Ops/s $\color{#35bf28}+0.22\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.2909ms 9.1988ms 108.7102 Ops/s 104.5624 Ops/s $\color{#35bf28}+3.97\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7595ms 1.5471ms 646.3769 Ops/s 644.2535 Ops/s $\color{#35bf28}+0.33\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4797ms 0.4237ms 2.3604 KOps/s 2.3331 KOps/s $\color{#35bf28}+1.17\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 34.8175ms 34.3089ms 29.1469 Ops/s 28.5469 Ops/s $\color{#35bf28}+2.10\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9391ms 1.7318ms 577.4283 Ops/s 573.5466 Ops/s $\color{#35bf28}+0.68\%$
test_dqn_speed[False-None] 1.5970ms 1.4156ms 706.4064 Ops/s 709.2552 Ops/s $\color{#d91a1a}-0.40\%$
test_dqn_speed[False-backward] 2.0272ms 1.9405ms 515.3192 Ops/s 515.0797 Ops/s $\color{#35bf28}+0.05\%$
test_dqn_speed[True-None] 0.6819ms 0.5397ms 1.8529 KOps/s 1.8285 KOps/s $\color{#35bf28}+1.33\%$
test_dqn_speed[True-backward] 1.0544ms 1.0030ms 997.0006 Ops/s 991.6164 Ops/s $\color{#35bf28}+0.54\%$
test_dqn_speed[reduce-overhead-None] 0.8709ms 0.5238ms 1.9093 KOps/s 1.8520 KOps/s $\color{#35bf28}+3.09\%$
test_ddpg_speed[False-None] 3.2294ms 2.8719ms 348.1969 Ops/s 345.8691 Ops/s $\color{#35bf28}+0.67\%$
test_ddpg_speed[False-backward] 4.2425ms 4.1193ms 242.7589 Ops/s 243.1405 Ops/s $\color{#d91a1a}-0.16\%$
test_ddpg_speed[True-None] 1.7843ms 1.4089ms 709.7883 Ops/s 710.1587 Ops/s $\color{#d91a1a}-0.05\%$
test_ddpg_speed[True-backward] 2.4833ms 2.4060ms 415.6287 Ops/s 343.0546 Ops/s $\textbf{\color{#35bf28}+21.16\%}$
test_ddpg_speed[reduce-overhead-None] 1.6714ms 1.3933ms 717.7450 Ops/s 674.7375 Ops/s $\textbf{\color{#35bf28}+6.37\%}$
test_sac_speed[False-None] 9.8808ms 8.1327ms 122.9598 Ops/s 123.0498 Ops/s $\color{#d91a1a}-0.07\%$
test_sac_speed[False-backward] 11.8362ms 11.4068ms 87.6667 Ops/s 87.9435 Ops/s $\color{#d91a1a}-0.31\%$
test_sac_speed[True-None] 2.5692ms 2.1667ms 461.5324 Ops/s 453.2615 Ops/s $\color{#35bf28}+1.82\%$
test_sac_speed[True-backward] 4.2153ms 4.0249ms 248.4561 Ops/s 245.0233 Ops/s $\color{#35bf28}+1.40\%$
test_sac_speed[reduce-overhead-None] 2.5431ms 2.1321ms 469.0159 Ops/s 452.8223 Ops/s $\color{#35bf28}+3.58\%$
test_redq_speed[False-None] 13.6875ms 10.6886ms 93.5579 Ops/s 96.1145 Ops/s $\color{#d91a1a}-2.66\%$
test_redq_speed[False-backward] 18.8056ms 18.0075ms 55.5326 Ops/s 55.8183 Ops/s $\color{#d91a1a}-0.51\%$
test_redq_speed[True-None] 4.9461ms 4.4920ms 222.6176 Ops/s 227.3021 Ops/s $\color{#d91a1a}-2.06\%$
test_redq_speed[True-backward] 10.0622ms 9.8452ms 101.5727 Ops/s 102.4228 Ops/s $\color{#d91a1a}-0.83\%$
test_redq_speed[reduce-overhead-None] 4.5808ms 4.4172ms 226.3875 Ops/s 219.2853 Ops/s $\color{#35bf28}+3.24\%$
test_redq_deprec_speed[False-None] 11.6788ms 11.1840ms 89.4137 Ops/s 88.4771 Ops/s $\color{#35bf28}+1.06\%$
test_redq_deprec_speed[False-backward] 16.4351ms 16.0999ms 62.1123 Ops/s 62.0273 Ops/s $\color{#35bf28}+0.14\%$
test_redq_deprec_speed[True-None] 4.1406ms 3.7108ms 269.4817 Ops/s 272.8290 Ops/s $\color{#d91a1a}-1.23\%$
test_redq_deprec_speed[True-backward] 8.2571ms 7.6839ms 130.1426 Ops/s 129.1197 Ops/s $\color{#35bf28}+0.79\%$
test_redq_deprec_speed[reduce-overhead-None] 3.7894ms 3.6183ms 276.3699 Ops/s 277.2159 Ops/s $\color{#d91a1a}-0.31\%$
test_td3_speed[False-None] 8.3025ms 8.1493ms 122.7098 Ops/s 123.1720 Ops/s $\color{#d91a1a}-0.38\%$
test_td3_speed[False-backward] 11.5078ms 11.0940ms 90.1386 Ops/s 90.1278 Ops/s $\color{#35bf28}+0.01\%$
test_td3_speed[True-None] 1.9726ms 1.8245ms 548.0934 Ops/s 546.6536 Ops/s $\color{#35bf28}+0.26\%$
test_td3_speed[True-backward] 3.7617ms 3.6655ms 272.8151 Ops/s 271.9752 Ops/s $\color{#35bf28}+0.31\%$
test_td3_speed[reduce-overhead-None] 2.4036ms 1.8127ms 551.6614 Ops/s 547.3763 Ops/s $\color{#35bf28}+0.78\%$
test_cql_speed[False-None] 30.1683ms 26.7553ms 37.3757 Ops/s 38.1134 Ops/s $\color{#d91a1a}-1.94\%$
test_cql_speed[False-backward] 38.6300ms 35.6141ms 28.0788 Ops/s 27.6900 Ops/s $\color{#35bf28}+1.40\%$
test_cql_speed[True-None] 13.1756ms 12.5905ms 79.4248 Ops/s 81.8360 Ops/s $\color{#d91a1a}-2.95\%$
test_cql_speed[True-backward] 19.0064ms 18.6317ms 53.6720 Ops/s 56.0518 Ops/s $\color{#d91a1a}-4.25\%$
test_cql_speed[reduce-overhead-None] 13.0477ms 12.5940ms 79.4031 Ops/s 77.1113 Ops/s $\color{#35bf28}+2.97\%$
test_a2c_speed[False-None] 5.7887ms 5.4859ms 182.2852 Ops/s 184.3495 Ops/s $\color{#d91a1a}-1.12\%$
test_a2c_speed[False-backward] 12.3300ms 12.0054ms 83.2956 Ops/s 83.7525 Ops/s $\color{#d91a1a}-0.55\%$
test_a2c_speed[True-None] 4.0699ms 3.7580ms 266.1019 Ops/s 271.9849 Ops/s $\color{#d91a1a}-2.16\%$
test_a2c_speed[True-backward] 8.9011ms 8.6727ms 115.3040 Ops/s 92.9059 Ops/s $\textbf{\color{#35bf28}+24.11\%}$
test_a2c_speed[reduce-overhead-None] 3.9428ms 3.6889ms 271.0829 Ops/s 270.0325 Ops/s $\color{#35bf28}+0.39\%$
test_ppo_speed[False-None] 6.2833ms 5.9621ms 167.7258 Ops/s 167.0988 Ops/s $\color{#35bf28}+0.38\%$
test_ppo_speed[False-backward] 13.3261ms 12.7208ms 78.6112 Ops/s 79.2572 Ops/s $\color{#d91a1a}-0.82\%$
test_ppo_speed[True-None] 3.7780ms 3.5882ms 278.6921 Ops/s 277.4676 Ops/s $\color{#35bf28}+0.44\%$
test_ppo_speed[True-backward] 8.7296ms 8.5339ms 117.1791 Ops/s 118.1938 Ops/s $\color{#d91a1a}-0.86\%$
test_ppo_speed[reduce-overhead-None] 3.8042ms 3.5858ms 278.8806 Ops/s 277.1466 Ops/s $\color{#35bf28}+0.63\%$
test_reinforce_speed[False-None] 4.7649ms 4.6141ms 216.7286 Ops/s 216.2492 Ops/s $\color{#35bf28}+0.22\%$
test_reinforce_speed[False-backward] 7.7824ms 7.4619ms 134.0137 Ops/s 134.7619 Ops/s $\color{#d91a1a}-0.56\%$
test_reinforce_speed[True-None] 3.0366ms 2.8693ms 348.5167 Ops/s 348.1842 Ops/s $\color{#35bf28}+0.10\%$
test_reinforce_speed[True-backward] 8.0854ms 7.8866ms 126.7977 Ops/s 128.1605 Ops/s $\color{#d91a1a}-1.06\%$
test_reinforce_speed[reduce-overhead-None] 2.9890ms 2.8566ms 350.0679 Ops/s 327.2593 Ops/s $\textbf{\color{#35bf28}+6.97\%}$
test_iql_speed[False-None] 20.7184ms 20.0434ms 49.8917 Ops/s 48.6271 Ops/s $\color{#35bf28}+2.60\%$
test_iql_speed[False-backward] 31.5050ms 30.7167ms 32.5555 Ops/s 32.4513 Ops/s $\color{#35bf28}+0.32\%$
test_iql_speed[True-None] 8.9058ms 8.5410ms 117.0822 Ops/s 115.4237 Ops/s $\color{#35bf28}+1.44\%$
test_iql_speed[True-backward] 17.0154ms 16.8259ms 59.4320 Ops/s 59.4395 Ops/s $\color{#d91a1a}-0.01\%$
test_iql_speed[reduce-overhead-None] 9.1607ms 8.5534ms 116.9125 Ops/s 112.8567 Ops/s $\color{#35bf28}+3.59\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2189ms 6.1036ms 163.8379 Ops/s 165.6569 Ops/s $\color{#d91a1a}-1.10\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7416s 0.8192ms 1.2207 KOps/s 3.1780 KOps/s $\textbf{\color{#d91a1a}-61.59\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5698ms 0.3296ms 3.0338 KOps/s 3.7540 KOps/s $\textbf{\color{#d91a1a}-19.18\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0836ms 5.8547ms 170.8022 Ops/s 171.7330 Ops/s $\color{#d91a1a}-0.54\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2845ms 0.3053ms 3.2755 KOps/s 3.4948 KOps/s $\textbf{\color{#d91a1a}-6.28\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5239ms 0.2846ms 3.5140 KOps/s 3.8237 KOps/s $\textbf{\color{#d91a1a}-8.10\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5637ms 1.3074ms 764.9007 Ops/s 762.4138 Ops/s $\color{#35bf28}+0.33\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.7555ms 1.2584ms 794.6420 Ops/s 830.1119 Ops/s $\color{#d91a1a}-4.27\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2518ms 6.0625ms 164.9490 Ops/s 166.7469 Ops/s $\color{#d91a1a}-1.08\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1631ms 0.4863ms 2.0564 KOps/s 1.9544 KOps/s $\textbf{\color{#35bf28}+5.22\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6548ms 0.4686ms 2.1339 KOps/s 2.1644 KOps/s $\color{#d91a1a}-1.41\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0048ms 5.9332ms 168.5439 Ops/s 170.2319 Ops/s $\color{#d91a1a}-0.99\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.1995ms 0.3419ms 2.9246 KOps/s 2.7237 KOps/s $\textbf{\color{#35bf28}+7.38\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6944ms 0.3351ms 2.9842 KOps/s 3.7339 KOps/s $\textbf{\color{#d91a1a}-20.08\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1436ms 5.8432ms 171.1396 Ops/s 170.0151 Ops/s $\color{#35bf28}+0.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0107ms 0.3277ms 3.0514 KOps/s 3.3278 KOps/s $\textbf{\color{#d91a1a}-8.31\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5196ms 0.3092ms 3.2338 KOps/s 3.2169 KOps/s $\color{#35bf28}+0.53\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1151ms 6.0323ms 165.7748 Ops/s 165.8674 Ops/s $\color{#d91a1a}-0.06\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.6683ms 0.4499ms 2.2226 KOps/s 1.9901 KOps/s $\textbf{\color{#35bf28}+11.68\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6089ms 0.4227ms 2.3658 KOps/s 2.1197 KOps/s $\textbf{\color{#35bf28}+11.61\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5503s 16.0003ms 62.4989 Ops/s 50.3068 Ops/s $\textbf{\color{#35bf28}+24.24\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 28.4887ms 2.4616ms 406.2327 Ops/s 508.9031 Ops/s $\textbf{\color{#d91a1a}-20.17\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.8403ms 1.1996ms 833.6247 Ops/s 816.1559 Ops/s $\color{#35bf28}+2.14\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.5112ms 5.0736ms 197.0988 Ops/s 196.4331 Ops/s $\color{#35bf28}+0.34\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 12.0917ms 1.9172ms 521.5923 Ops/s 523.2893 Ops/s $\color{#d91a1a}-0.32\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.9827ms 0.8803ms 1.1359 KOps/s 1.1503 KOps/s $\color{#d91a1a}-1.25\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5157s 15.5071ms 64.4865 Ops/s 56.4119 Ops/s $\textbf{\color{#35bf28}+14.31\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 12.8629ms 2.2447ms 445.4981 Ops/s 458.2169 Ops/s $\color{#d91a1a}-2.78\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.3793ms 1.2155ms 822.7197 Ops/s 929.9363 Ops/s $\textbf{\color{#d91a1a}-11.53\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.5799ms 35.8567ms 27.8888 Ops/s 27.8494 Ops/s $\color{#35bf28}+0.14\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.4145ms 18.3508ms 54.4936 Ops/s 55.0502 Ops/s $\color{#d91a1a}-1.01\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 40.2604ms 36.7954ms 27.1773 Ops/s 26.9187 Ops/s $\color{#35bf28}+0.96\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.9699ms 18.5230ms 53.9870 Ops/s 52.3523 Ops/s $\color{#35bf28}+3.12\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 39.9868ms 38.5382ms 25.9483 Ops/s 25.6679 Ops/s $\color{#35bf28}+1.09\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.4193ms 19.9291ms 50.1779 Ops/s 49.8847 Ops/s $\color{#35bf28}+0.59\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CI Has to do with CI setup (e.g. wheels & builds, tests...) 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