+
Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 21, 2024

No description provided.

Vincent Moens added 2 commits October 21, 2024 14:25
Copy link

pytorch-bot bot commented Oct 21, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2508

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

❌ 2 New Failures, 6 Unrelated Failures

As of commit 9d93f72 with merge base 56b0b9a (image):

NEW FAILURES - The following jobs have failed:

FLAKY - The following job failed but was likely due to flakiness present on trunk:

BROKEN TRUNK - The following jobs failed but was present on the merge base:

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

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

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Oct 21, 2024
@vmoens vmoens added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Oct 21, 2024
Copy link

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}2$. Worsened: $\large\color{#d91a1a}20$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4212s 0.4181s 2.3920 Ops/s 2.2750 Ops/s $\textbf{\color{#35bf28}+5.14\%}$
test_transformed 0.7075s 0.6153s 1.6252 Ops/s 1.6849 Ops/s $\color{#d91a1a}-3.54\%$
test_serial 1.4467s 1.3558s 0.7376 Ops/s 0.7322 Ops/s $\color{#35bf28}+0.73\%$
test_parallel 1.4254s 1.3379s 0.7474 Ops/s 0.7412 Ops/s $\color{#35bf28}+0.84\%$
test_step_mdp_speed[True-True-True-True-True] 0.2449ms 29.0004μs 34.4823 KOps/s 34.8037 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[True-True-True-True-False] 95.9000μs 17.7937μs 56.1997 KOps/s 58.7654 KOps/s $\color{#d91a1a}-4.37\%$
test_step_mdp_speed[True-True-True-False-True] 43.8020μs 15.9339μs 62.7594 KOps/s 63.0124 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[True-True-True-False-False] 58.9200μs 9.6494μs 103.6336 KOps/s 107.1143 KOps/s $\color{#d91a1a}-3.25\%$
test_step_mdp_speed[True-True-False-True-True] 66.5040μs 31.3405μs 31.9076 KOps/s 32.6591 KOps/s $\color{#d91a1a}-2.30\%$
test_step_mdp_speed[True-True-False-True-False] 68.7390μs 19.8329μs 50.4213 KOps/s 51.9009 KOps/s $\color{#d91a1a}-2.85\%$
test_step_mdp_speed[True-True-False-False-True] 60.4230μs 18.1810μs 55.0025 KOps/s 56.9846 KOps/s $\color{#d91a1a}-3.48\%$
test_step_mdp_speed[True-True-False-False-False] 51.5960μs 11.8480μs 84.4026 KOps/s 87.5504 KOps/s $\color{#d91a1a}-3.60\%$
test_step_mdp_speed[True-False-True-True-True] 83.2460μs 33.6480μs 29.7195 KOps/s 30.6340 KOps/s $\color{#d91a1a}-2.99\%$
test_step_mdp_speed[True-False-True-True-False] 73.0760μs 22.0103μs 45.4332 KOps/s 47.3575 KOps/s $\color{#d91a1a}-4.06\%$
test_step_mdp_speed[True-False-True-False-True] 49.5220μs 18.2005μs 54.9435 KOps/s 56.0421 KOps/s $\color{#d91a1a}-1.96\%$
test_step_mdp_speed[True-False-True-False-False] 57.0760μs 11.7069μs 85.4196 KOps/s 88.3002 KOps/s $\color{#d91a1a}-3.26\%$
test_step_mdp_speed[True-False-False-True-True] 65.7530μs 35.6101μs 28.0819 KOps/s 29.2282 KOps/s $\color{#d91a1a}-3.92\%$
test_step_mdp_speed[True-False-False-True-False] 53.6410μs 23.9957μs 41.6741 KOps/s 43.5259 KOps/s $\color{#d91a1a}-4.25\%$
test_step_mdp_speed[True-False-False-False-True] 59.7320μs 20.3873μs 49.0500 KOps/s 50.4815 KOps/s $\color{#d91a1a}-2.84\%$
test_step_mdp_speed[True-False-False-False-False] 42.7800μs 13.7390μs 72.7853 KOps/s 74.7778 KOps/s $\color{#d91a1a}-2.66\%$
test_step_mdp_speed[False-True-True-True-True] 87.5440μs 33.5231μs 29.8302 KOps/s 30.5637 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[False-True-True-True-False] 68.1370μs 22.0219μs 45.4093 KOps/s 47.5018 KOps/s $\color{#d91a1a}-4.40\%$
test_step_mdp_speed[False-True-True-False-True] 65.1910μs 21.3693μs 46.7960 KOps/s 47.7288 KOps/s $\color{#d91a1a}-1.95\%$
test_step_mdp_speed[False-True-True-False-False] 60.6330μs 13.4469μs 74.3665 KOps/s 76.2740 KOps/s $\color{#d91a1a}-2.50\%$
test_step_mdp_speed[False-True-False-True-True] 80.2800μs 35.2302μs 28.3847 KOps/s 29.0922 KOps/s $\color{#d91a1a}-2.43\%$
test_step_mdp_speed[False-True-False-True-False] 58.9610μs 24.1744μs 41.3660 KOps/s 43.6480 KOps/s $\textbf{\color{#d91a1a}-5.23\%}$
test_step_mdp_speed[False-True-False-False-True] 2.7166ms 23.6500μs 42.2832 KOps/s 43.6374 KOps/s $\color{#d91a1a}-3.10\%$
test_step_mdp_speed[False-True-False-False-False] 61.5450μs 15.7508μs 63.4890 KOps/s 66.3159 KOps/s $\color{#d91a1a}-4.26\%$
test_step_mdp_speed[False-False-True-True-True] 69.7590μs 37.6085μs 26.5898 KOps/s 27.6278 KOps/s $\color{#d91a1a}-3.76\%$
test_step_mdp_speed[False-False-True-True-False] 70.5810μs 26.3069μs 38.0129 KOps/s 39.8298 KOps/s $\color{#d91a1a}-4.56\%$
test_step_mdp_speed[False-False-True-False-True] 48.3500μs 23.5384μs 42.4838 KOps/s 43.1493 KOps/s $\color{#d91a1a}-1.54\%$
test_step_mdp_speed[False-False-True-False-False] 59.9520μs 15.6002μs 64.1019 KOps/s 66.0817 KOps/s $\color{#d91a1a}-3.00\%$
test_step_mdp_speed[False-False-False-True-True] 88.3250μs 39.3772μs 25.3954 KOps/s 26.4578 KOps/s $\color{#d91a1a}-4.02\%$
test_step_mdp_speed[False-False-False-True-False] 69.5100μs 28.0720μs 35.6226 KOps/s 37.4358 KOps/s $\color{#d91a1a}-4.84\%$
test_step_mdp_speed[False-False-False-False-True] 58.8200μs 25.4848μs 39.2391 KOps/s 38.1409 KOps/s $\color{#35bf28}+2.88\%$
test_step_mdp_speed[False-False-False-False-False] 50.7860μs 17.6100μs 56.7858 KOps/s 58.0869 KOps/s $\color{#d91a1a}-2.24\%$
test_values[generalized_advantage_estimate-True-True] 15.2579ms 10.1725ms 98.3044 Ops/s 101.5720 Ops/s $\color{#d91a1a}-3.22\%$
test_values[vec_generalized_advantage_estimate-True-True] 39.1026ms 35.3847ms 28.2608 Ops/s 29.5930 Ops/s $\color{#d91a1a}-4.50\%$
test_values[td0_return_estimate-False-False] 0.2494ms 0.1910ms 5.2362 KOps/s 5.1710 KOps/s $\color{#35bf28}+1.26\%$
test_values[td1_return_estimate-False-False] 28.2393ms 23.9150ms 41.8147 Ops/s 40.4977 Ops/s $\color{#35bf28}+3.25\%$
test_values[vec_td1_return_estimate-False-False] 37.8907ms 35.6740ms 28.0316 Ops/s 29.4343 Ops/s $\color{#d91a1a}-4.77\%$
test_values[td_lambda_return_estimate-True-False] 34.1748ms 33.7294ms 29.6477 Ops/s 28.3710 Ops/s $\color{#35bf28}+4.50\%$
test_values[vec_td_lambda_return_estimate-True-False] 41.0526ms 35.7154ms 27.9991 Ops/s 29.4797 Ops/s $\textbf{\color{#d91a1a}-5.02\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.6707ms 8.4299ms 118.6248 Ops/s 118.4137 Ops/s $\color{#35bf28}+0.18\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.7139ms 2.0529ms 487.1043 Ops/s 493.6166 Ops/s $\color{#d91a1a}-1.32\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5016ms 0.3598ms 2.7791 KOps/s 2.7337 KOps/s $\color{#35bf28}+1.66\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 49.7850ms 47.7722ms 20.9327 Ops/s 22.9340 Ops/s $\textbf{\color{#d91a1a}-8.73\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.0284ms 3.0655ms 326.2146 Ops/s 319.9916 Ops/s $\color{#35bf28}+1.94\%$
test_dqn_speed[False-None] 6.1148ms 1.3635ms 733.4282 Ops/s 724.3248 Ops/s $\color{#35bf28}+1.26\%$
test_dqn_speed[False-backward] 2.6683ms 1.8748ms 533.3886 Ops/s 535.5306 Ops/s $\color{#d91a1a}-0.40\%$
test_dqn_speed[True-None] 0.6919ms 0.4688ms 2.1330 KOps/s 2.1118 KOps/s $\color{#35bf28}+1.00\%$
test_dqn_speed[True-backward] 0.9428ms 0.8919ms 1.1212 KOps/s 1.0848 KOps/s $\color{#35bf28}+3.36\%$
test_dqn_speed[reduce-overhead-None] 0.8100ms 0.4726ms 2.1160 KOps/s 2.0459 KOps/s $\color{#35bf28}+3.43\%$
test_dqn_speed[reduce-overhead-backward] 0.9329ms 0.8882ms 1.1258 KOps/s 1.0976 KOps/s $\color{#35bf28}+2.57\%$
test_ddpg_speed[False-None] 3.5924ms 2.8199ms 354.6230 Ops/s 343.4439 Ops/s $\color{#35bf28}+3.26\%$
test_ddpg_speed[False-backward] 4.9835ms 4.0323ms 247.9959 Ops/s 241.4570 Ops/s $\color{#35bf28}+2.71\%$
test_ddpg_speed[True-None] 1.2149ms 1.0096ms 990.4536 Ops/s 967.4440 Ops/s $\color{#35bf28}+2.38\%$
test_ddpg_speed[True-backward] 2.0113ms 1.9210ms 520.5566 Ops/s 508.2219 Ops/s $\color{#35bf28}+2.43\%$
test_ddpg_speed[reduce-overhead-None] 1.4631ms 1.0186ms 981.7840 Ops/s 979.5917 Ops/s $\color{#35bf28}+0.22\%$
test_ddpg_speed[reduce-overhead-backward] 2.0607ms 1.9647ms 508.9746 Ops/s 514.5444 Ops/s $\color{#d91a1a}-1.08\%$
test_sac_speed[False-None] 10.0102ms 8.1958ms 122.0135 Ops/s 123.8084 Ops/s $\color{#d91a1a}-1.45\%$
test_sac_speed[False-backward] 11.7877ms 10.9062ms 91.6907 Ops/s 92.2702 Ops/s $\color{#d91a1a}-0.63\%$
test_sac_speed[True-None] 2.6391ms 1.9980ms 500.4900 Ops/s 529.2783 Ops/s $\textbf{\color{#d91a1a}-5.44\%}$
test_sac_speed[True-backward] 3.7081ms 3.5843ms 278.9919 Ops/s 269.0800 Ops/s $\color{#35bf28}+3.68\%$
test_sac_speed[reduce-overhead-None] 3.9747ms 1.8770ms 532.7568 Ops/s 530.4749 Ops/s $\color{#35bf28}+0.43\%$
test_sac_speed[reduce-overhead-backward] 4.4332ms 3.6619ms 273.0795 Ops/s 275.0077 Ops/s $\color{#d91a1a}-0.70\%$
test_redq_speed[False-None] 14.8390ms 13.1861ms 75.8375 Ops/s 75.6714 Ops/s $\color{#35bf28}+0.22\%$
test_redq_speed[False-backward] 24.1178ms 22.7396ms 43.9762 Ops/s 44.0945 Ops/s $\color{#d91a1a}-0.27\%$
test_redq_speed[True-None] 5.9214ms 5.1077ms 195.7843 Ops/s 197.0747 Ops/s $\color{#d91a1a}-0.65\%$
test_redq_speed[True-backward] 13.3547ms 12.7097ms 78.6799 Ops/s 78.9239 Ops/s $\color{#d91a1a}-0.31\%$
test_redq_speed[reduce-overhead-None] 6.4833ms 5.6089ms 178.2866 Ops/s 198.7455 Ops/s $\textbf{\color{#d91a1a}-10.29\%}$
test_redq_speed[reduce-overhead-backward] 13.7106ms 13.1923ms 75.8019 Ops/s 78.6508 Ops/s $\color{#d91a1a}-3.62\%$
test_redq_deprec_speed[False-None] 16.8435ms 14.2723ms 70.0658 Ops/s 73.9563 Ops/s $\textbf{\color{#d91a1a}-5.26\%}$
test_redq_deprec_speed[False-backward] 21.5913ms 20.4567ms 48.8838 Ops/s 51.3620 Ops/s $\color{#d91a1a}-4.82\%$
test_redq_deprec_speed[True-None] 5.0628ms 4.2989ms 232.6182 Ops/s 260.9837 Ops/s $\textbf{\color{#d91a1a}-10.87\%}$
test_redq_deprec_speed[True-backward] 9.4703ms 9.2232ms 108.4224 Ops/s 115.4738 Ops/s $\textbf{\color{#d91a1a}-6.11\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.8949ms 4.1691ms 239.8605 Ops/s 264.8905 Ops/s $\textbf{\color{#d91a1a}-9.45\%}$
test_redq_deprec_speed[reduce-overhead-backward] 9.8449ms 8.9266ms 112.0251 Ops/s 115.6960 Ops/s $\color{#d91a1a}-3.17\%$
test_td3_speed[False-None] 10.2428ms 8.2865ms 120.6779 Ops/s 121.6570 Ops/s $\color{#d91a1a}-0.80\%$
test_td3_speed[False-backward] 13.2572ms 10.7212ms 93.2727 Ops/s 90.3101 Ops/s $\color{#35bf28}+3.28\%$
test_td3_speed[True-None] 2.0954ms 1.8075ms 553.2515 Ops/s 564.9059 Ops/s $\color{#d91a1a}-2.06\%$
test_td3_speed[True-backward] 3.7046ms 3.5639ms 280.5922 Ops/s 274.4033 Ops/s $\color{#35bf28}+2.26\%$
test_td3_speed[reduce-overhead-None] 2.0057ms 1.7745ms 563.5279 Ops/s 547.8341 Ops/s $\color{#35bf28}+2.86\%$
test_td3_speed[reduce-overhead-backward] 3.7811ms 3.5216ms 283.9640 Ops/s 284.3450 Ops/s $\color{#d91a1a}-0.13\%$
test_cql_speed[False-None] 39.7708ms 36.6837ms 27.2601 Ops/s 27.4522 Ops/s $\color{#d91a1a}-0.70\%$
test_cql_speed[False-backward] 50.7723ms 47.8401ms 20.9029 Ops/s 21.4955 Ops/s $\color{#d91a1a}-2.76\%$
test_cql_speed[True-None] 16.9326ms 16.0918ms 62.1435 Ops/s 62.6023 Ops/s $\color{#d91a1a}-0.73\%$
test_cql_speed[True-backward] 24.3297ms 22.8039ms 43.8522 Ops/s 42.7419 Ops/s $\color{#35bf28}+2.60\%$
test_cql_speed[reduce-overhead-None] 17.1094ms 15.8791ms 62.9759 Ops/s 62.2393 Ops/s $\color{#35bf28}+1.18\%$
test_cql_speed[reduce-overhead-backward] 24.3538ms 23.2852ms 42.9457 Ops/s 42.6083 Ops/s $\color{#35bf28}+0.79\%$
test_a2c_speed[False-None] 8.6954ms 7.5727ms 132.0530 Ops/s 133.5812 Ops/s $\color{#d91a1a}-1.14\%$
test_a2c_speed[False-backward] 15.5767ms 14.8254ms 67.4517 Ops/s 66.5665 Ops/s $\color{#35bf28}+1.33\%$
test_a2c_speed[True-None] 4.2605ms 3.4253ms 291.9464 Ops/s 289.0894 Ops/s $\color{#35bf28}+0.99\%$
test_a2c_speed[True-backward] 11.0500ms 10.7184ms 93.2975 Ops/s 96.1599 Ops/s $\color{#d91a1a}-2.98\%$
test_a2c_speed[reduce-overhead-None] 3.8062ms 3.4941ms 286.1953 Ops/s 288.5248 Ops/s $\color{#d91a1a}-0.81\%$
test_a2c_speed[reduce-overhead-backward] 11.8806ms 10.5669ms 94.6355 Ops/s 96.3484 Ops/s $\color{#d91a1a}-1.78\%$
test_ppo_speed[False-None] 8.4546ms 7.8653ms 127.1405 Ops/s 127.6959 Ops/s $\color{#d91a1a}-0.43\%$
test_ppo_speed[False-backward] 16.9017ms 15.3935ms 64.9626 Ops/s 66.0346 Ops/s $\color{#d91a1a}-1.62\%$
test_ppo_speed[True-None] 4.2419ms 3.8755ms 258.0284 Ops/s 259.2632 Ops/s $\color{#d91a1a}-0.48\%$
test_ppo_speed[True-backward] 10.7663ms 10.2336ms 97.7170 Ops/s 98.0036 Ops/s $\color{#d91a1a}-0.29\%$
test_ppo_speed[reduce-overhead-None] 4.4951ms 3.8095ms 262.5019 Ops/s 263.4664 Ops/s $\color{#d91a1a}-0.37\%$
test_ppo_speed[reduce-overhead-backward] 10.8909ms 10.1635ms 98.3913 Ops/s 97.8382 Ops/s $\color{#35bf28}+0.57\%$
test_reinforce_speed[False-None] 9.2380ms 6.6714ms 149.8930 Ops/s 151.4433 Ops/s $\color{#d91a1a}-1.02\%$
test_reinforce_speed[False-backward] 10.3520ms 10.0909ms 99.0991 Ops/s 99.5663 Ops/s $\color{#d91a1a}-0.47\%$
test_reinforce_speed[True-None] 3.3758ms 2.7799ms 359.7203 Ops/s 367.0300 Ops/s $\color{#d91a1a}-1.99\%$
test_reinforce_speed[True-backward] 9.9503ms 9.5464ms 104.7517 Ops/s 110.9084 Ops/s $\textbf{\color{#d91a1a}-5.55\%}$
test_reinforce_speed[reduce-overhead-None] 3.6484ms 2.9321ms 341.0515 Ops/s 349.2373 Ops/s $\color{#d91a1a}-2.34\%$
test_reinforce_speed[reduce-overhead-backward] 10.1195ms 9.4342ms 105.9976 Ops/s 109.5950 Ops/s $\color{#d91a1a}-3.28\%$
test_iql_speed[False-None] 34.3960ms 32.9036ms 30.3918 Ops/s 30.1380 Ops/s $\color{#35bf28}+0.84\%$
test_iql_speed[False-backward] 59.6946ms 46.8269ms 21.3552 Ops/s 21.5883 Ops/s $\color{#d91a1a}-1.08\%$
test_iql_speed[True-None] 11.9185ms 11.1463ms 89.7157 Ops/s 89.9322 Ops/s $\color{#d91a1a}-0.24\%$
test_iql_speed[True-backward] 24.7661ms 22.9406ms 43.5909 Ops/s 43.0103 Ops/s $\color{#35bf28}+1.35\%$
test_iql_speed[reduce-overhead-None] 11.7346ms 11.1415ms 89.7542 Ops/s 89.2358 Ops/s $\color{#35bf28}+0.58\%$
test_iql_speed[reduce-overhead-backward] 24.1379ms 23.1986ms 43.1060 Ops/s 43.5490 Ops/s $\color{#d91a1a}-1.02\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.2706ms 5.3417ms 187.2060 Ops/s 193.1559 Ops/s $\color{#d91a1a}-3.08\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7726ms 0.5060ms 1.9762 KOps/s 2.0149 KOps/s $\color{#d91a1a}-1.92\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8343ms 0.4784ms 2.0902 KOps/s 2.1503 KOps/s $\color{#d91a1a}-2.80\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.8324ms 5.0552ms 197.8146 Ops/s 204.8198 Ops/s $\color{#d91a1a}-3.42\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 3.0700ms 0.5011ms 1.9957 KOps/s 2.0323 KOps/s $\color{#d91a1a}-1.80\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6934ms 0.4729ms 2.1147 KOps/s 2.1504 KOps/s $\color{#d91a1a}-1.66\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.3494ms 1.6159ms 618.8428 Ops/s 620.7171 Ops/s $\color{#d91a1a}-0.30\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.1904ms 1.5612ms 640.5153 Ops/s 636.1644 Ops/s $\color{#35bf28}+0.68\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.6309ms 5.2699ms 189.7578 Ops/s 196.2382 Ops/s $\color{#d91a1a}-3.30\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2555ms 0.6468ms 1.5462 KOps/s 1.5931 KOps/s $\color{#d91a1a}-2.95\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8503ms 0.6110ms 1.6367 KOps/s 1.6491 KOps/s $\color{#d91a1a}-0.75\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7109ms 5.2018ms 192.2418 Ops/s 203.3912 Ops/s $\textbf{\color{#d91a1a}-5.48\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1413ms 0.5402ms 1.8510 KOps/s 1.9592 KOps/s $\textbf{\color{#d91a1a}-5.52\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6931ms 0.4741ms 2.1091 KOps/s 2.0942 KOps/s $\color{#35bf28}+0.71\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.5513ms 5.0885ms 196.5229 Ops/s 200.4024 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.1928ms 0.5014ms 1.9944 KOps/s 2.0071 KOps/s $\color{#d91a1a}-0.63\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7175ms 0.4765ms 2.0986 KOps/s 2.0821 KOps/s $\color{#35bf28}+0.79\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 8.1351ms 5.5909ms 178.8619 Ops/s 188.3415 Ops/s $\textbf{\color{#d91a1a}-5.03\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.7614ms 0.6496ms 1.5393 KOps/s 1.5663 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8589ms 0.6098ms 1.6399 KOps/s 1.6187 KOps/s $\color{#35bf28}+1.31\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4849ms 4.7050ms 212.5413 Ops/s 233.8966 Ops/s $\textbf{\color{#d91a1a}-9.13\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.1046ms 2.4171ms 413.7177 Ops/s 441.7822 Ops/s $\textbf{\color{#d91a1a}-6.35\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.2063ms 1.3391ms 746.7575 Ops/s 792.5968 Ops/s $\textbf{\color{#d91a1a}-5.78\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4959s 14.3944ms 69.4712 Ops/s 229.5421 Ops/s $\textbf{\color{#d91a1a}-69.73\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.8302ms 2.2598ms 442.5260 Ops/s 398.5036 Ops/s $\textbf{\color{#35bf28}+11.05\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1816ms 1.4554ms 687.0929 Ops/s 807.6214 Ops/s $\textbf{\color{#d91a1a}-14.92\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.8640ms 4.9195ms 203.2710 Ops/s 221.7612 Ops/s $\textbf{\color{#d91a1a}-8.34\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.0909ms 2.5918ms 385.8270 Ops/s 415.1423 Ops/s $\textbf{\color{#d91a1a}-7.06\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.6563ms 1.5074ms 663.3878 Ops/s 681.5065 Ops/s $\color{#d91a1a}-2.66\%$

Copy link

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}14$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7068s 0.7030s 1.4225 Ops/s 1.4119 Ops/s $\color{#35bf28}+0.76\%$
test_transformed 1.0368s 0.9616s 1.0400 Ops/s 1.0424 Ops/s $\color{#d91a1a}-0.23\%$
test_serial 2.1563s 2.0777s 0.4813 Ops/s 0.4866 Ops/s $\color{#d91a1a}-1.08\%$
test_parallel 2.0462s 1.9861s 0.5035 Ops/s 0.5129 Ops/s $\color{#d91a1a}-1.83\%$
test_step_mdp_speed[True-True-True-True-True] 0.2478ms 38.7878μs 25.7813 KOps/s 27.1124 KOps/s $\color{#d91a1a}-4.91\%$
test_step_mdp_speed[True-True-True-True-False] 0.4125ms 22.2316μs 44.9811 KOps/s 45.1568 KOps/s $\color{#d91a1a}-0.39\%$
test_step_mdp_speed[True-True-True-False-True] 0.4170ms 21.0561μs 47.4921 KOps/s 50.0586 KOps/s $\textbf{\color{#d91a1a}-5.13\%}$
test_step_mdp_speed[True-True-True-False-False] 71.1110μs 12.2389μs 81.7070 KOps/s 83.2730 KOps/s $\color{#d91a1a}-1.88\%$
test_step_mdp_speed[True-True-False-True-True] 0.4247ms 41.6363μs 24.0175 KOps/s 25.6251 KOps/s $\textbf{\color{#d91a1a}-6.27\%}$
test_step_mdp_speed[True-True-False-True-False] 0.4083ms 25.1252μs 39.8006 KOps/s 40.7873 KOps/s $\color{#d91a1a}-2.42\%$
test_step_mdp_speed[True-True-False-False-True] 62.2210μs 23.8951μs 41.8496 KOps/s 44.2944 KOps/s $\textbf{\color{#d91a1a}-5.52\%}$
test_step_mdp_speed[True-True-False-False-False] 49.5700μs 14.9964μs 66.6827 KOps/s 69.7034 KOps/s $\color{#d91a1a}-4.33\%$
test_step_mdp_speed[True-False-True-True-True] 0.4367ms 44.5136μs 22.4650 KOps/s 23.7270 KOps/s $\textbf{\color{#d91a1a}-5.32\%}$
test_step_mdp_speed[True-False-True-True-False] 0.4091ms 27.3665μs 36.5410 KOps/s 36.9466 KOps/s $\color{#d91a1a}-1.10\%$
test_step_mdp_speed[True-False-True-False-True] 83.0810μs 23.5671μs 42.4321 KOps/s 42.4558 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[True-False-True-False-False] 0.4193ms 14.8495μs 67.3424 KOps/s 69.4937 KOps/s $\color{#d91a1a}-3.10\%$
test_step_mdp_speed[True-False-False-True-True] 0.4301ms 46.0639μs 21.7090 KOps/s 22.4023 KOps/s $\color{#d91a1a}-3.10\%$
test_step_mdp_speed[True-False-False-True-False] 0.4090ms 29.5964μs 33.7879 KOps/s 33.6747 KOps/s $\color{#35bf28}+0.34\%$
test_step_mdp_speed[True-False-False-False-True] 61.6600μs 26.1289μs 38.2718 KOps/s 40.5974 KOps/s $\textbf{\color{#d91a1a}-5.73\%}$
test_step_mdp_speed[True-False-False-False-False] 0.4023ms 17.3108μs 57.7674 KOps/s 59.5955 KOps/s $\color{#d91a1a}-3.07\%$
test_step_mdp_speed[False-True-True-True-True] 0.4394ms 43.4561μs 23.0117 KOps/s 23.9288 KOps/s $\color{#d91a1a}-3.83\%$
test_step_mdp_speed[False-True-True-True-False] 69.7110μs 27.0947μs 36.9076 KOps/s 37.6749 KOps/s $\color{#d91a1a}-2.04\%$
test_step_mdp_speed[False-True-True-False-True] 0.4156ms 28.5699μs 35.0019 KOps/s 37.3116 KOps/s $\textbf{\color{#d91a1a}-6.19\%}$
test_step_mdp_speed[False-True-True-False-False] 0.3951ms 16.9817μs 58.8870 KOps/s 59.9437 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[False-True-False-True-True] 0.4280ms 46.4862μs 21.5118 KOps/s 22.3335 KOps/s $\color{#d91a1a}-3.68\%$
test_step_mdp_speed[False-True-False-True-False] 0.4211ms 29.5474μs 33.8439 KOps/s 33.7024 KOps/s $\color{#35bf28}+0.42\%$
test_step_mdp_speed[False-True-False-False-True] 3.3764ms 31.1498μs 32.1029 KOps/s 33.5645 KOps/s $\color{#d91a1a}-4.35\%$
test_step_mdp_speed[False-True-False-False-False] 0.4162ms 19.3681μs 51.6313 KOps/s 51.3945 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[False-False-True-True-True] 0.4294ms 48.9995μs 20.4084 KOps/s 21.0970 KOps/s $\color{#d91a1a}-3.26\%$
test_step_mdp_speed[False-False-True-True-False] 0.4153ms 32.4147μs 30.8502 KOps/s 31.1211 KOps/s $\color{#d91a1a}-0.87\%$
test_step_mdp_speed[False-False-True-False-True] 0.1604ms 30.1848μs 33.1292 KOps/s 33.1887 KOps/s $\color{#d91a1a}-0.18\%$
test_step_mdp_speed[False-False-True-False-False] 0.4074ms 19.1401μs 52.2463 KOps/s 51.4658 KOps/s $\color{#35bf28}+1.52\%$
test_step_mdp_speed[False-False-False-True-True] 0.4283ms 50.4285μs 19.8301 KOps/s 20.3801 KOps/s $\color{#d91a1a}-2.70\%$
test_step_mdp_speed[False-False-False-True-False] 0.4419ms 34.7316μs 28.7923 KOps/s 28.9105 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[False-False-False-False-True] 68.2210μs 32.5776μs 30.6959 KOps/s 31.1407 KOps/s $\color{#d91a1a}-1.43\%$
test_step_mdp_speed[False-False-False-False-False] 0.3995ms 21.9599μs 45.5376 KOps/s 46.3738 KOps/s $\color{#d91a1a}-1.80\%$
test_values[generalized_advantage_estimate-True-True] 23.3650ms 22.7567ms 43.9431 Ops/s 43.6696 Ops/s $\color{#35bf28}+0.63\%$
test_values[vec_generalized_advantage_estimate-True-True] 92.2377ms 2.7093ms 369.1012 Ops/s 341.9571 Ops/s $\textbf{\color{#35bf28}+7.94\%}$
test_values[td0_return_estimate-False-False] 83.0210μs 62.2690μs 16.0594 KOps/s 15.8046 KOps/s $\color{#35bf28}+1.61\%$
test_values[td1_return_estimate-False-False] 51.7011ms 50.7442ms 19.7067 Ops/s 19.6073 Ops/s $\color{#35bf28}+0.51\%$
test_values[vec_td1_return_estimate-False-False] 1.3574ms 1.0393ms 962.1476 Ops/s 961.3509 Ops/s $\color{#35bf28}+0.08\%$
test_values[td_lambda_return_estimate-True-False] 82.4248ms 81.2700ms 12.3047 Ops/s 12.3225 Ops/s $\color{#d91a1a}-0.14\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2950ms 1.0301ms 970.7647 Ops/s 961.8109 Ops/s $\color{#35bf28}+0.93\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 23.1232ms 22.5369ms 44.3716 Ops/s 43.9661 Ops/s $\color{#35bf28}+0.92\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0045ms 0.7062ms 1.4160 KOps/s 1.4052 KOps/s $\color{#35bf28}+0.77\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7808ms 0.6223ms 1.6069 KOps/s 1.5838 KOps/s $\color{#35bf28}+1.46\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.8173ms 1.4352ms 696.7625 Ops/s 694.0106 Ops/s $\color{#35bf28}+0.40\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.0257ms 0.6381ms 1.5671 KOps/s 1.5550 KOps/s $\color{#35bf28}+0.78\%$
test_dqn_speed[False-None] 7.1953ms 1.3123ms 762.0433 Ops/s 750.9374 Ops/s $\color{#35bf28}+1.48\%$
test_dqn_speed[False-backward] 2.3015ms 1.8478ms 541.1980 Ops/s 538.5313 Ops/s $\color{#35bf28}+0.50\%$
test_dqn_speed[True-None] 0.7455ms 0.5639ms 1.7735 KOps/s 1.7910 KOps/s $\color{#d91a1a}-0.98\%$
test_dqn_speed[True-backward] 1.2055ms 1.0037ms 996.3424 Ops/s 818.5180 Ops/s $\textbf{\color{#35bf28}+21.73\%}$
test_dqn_speed[reduce-overhead-None] 0.9353ms 0.5569ms 1.7956 KOps/s 1.7898 KOps/s $\color{#35bf28}+0.32\%$
test_dqn_speed[reduce-overhead-backward] 1.4110ms 1.0035ms 996.4887 Ops/s 990.0940 Ops/s $\color{#35bf28}+0.65\%$
test_ddpg_speed[False-None] 2.8648ms 2.6648ms 375.2568 Ops/s 370.4708 Ops/s $\color{#35bf28}+1.29\%$
test_ddpg_speed[False-backward] 4.0265ms 3.8719ms 258.2689 Ops/s 254.5345 Ops/s $\color{#35bf28}+1.47\%$
test_ddpg_speed[True-None] 1.6498ms 1.2306ms 812.6311 Ops/s 811.0116 Ops/s $\color{#35bf28}+0.20\%$
test_ddpg_speed[True-backward] 2.5970ms 2.2482ms 444.8070 Ops/s 414.9171 Ops/s $\textbf{\color{#35bf28}+7.20\%}$
test_ddpg_speed[reduce-overhead-None] 1.6526ms 1.2470ms 801.9457 Ops/s 798.7540 Ops/s $\color{#35bf28}+0.40\%$
test_ddpg_speed[reduce-overhead-backward] 2.4293ms 2.2305ms 448.3271 Ops/s 448.4789 Ops/s $\color{#d91a1a}-0.03\%$
test_sac_speed[False-None] 8.0713ms 7.4998ms 133.3367 Ops/s 130.6485 Ops/s $\color{#35bf28}+2.06\%$
test_sac_speed[False-backward] 11.4568ms 10.7650ms 92.8934 Ops/s 91.7309 Ops/s $\color{#35bf28}+1.27\%$
test_sac_speed[True-None] 2.4658ms 2.0310ms 492.3721 Ops/s 492.8106 Ops/s $\color{#d91a1a}-0.09\%$
test_sac_speed[True-backward] 4.2347ms 3.9553ms 252.8225 Ops/s 216.5254 Ops/s $\textbf{\color{#35bf28}+16.76\%}$
test_sac_speed[reduce-overhead-None] 2.4371ms 2.0150ms 496.2821 Ops/s 487.0622 Ops/s $\color{#35bf28}+1.89\%$
test_sac_speed[reduce-overhead-backward] 4.2304ms 3.9708ms 251.8378 Ops/s 248.1786 Ops/s $\color{#35bf28}+1.47\%$
test_redq_speed[False-None] 14.9665ms 10.3212ms 96.8880 Ops/s 67.0156 Ops/s $\textbf{\color{#35bf28}+44.58\%}$
test_redq_speed[False-backward] 18.4747ms 17.5494ms 56.9818 Ops/s 56.9199 Ops/s $\color{#35bf28}+0.11\%$
test_redq_speed[True-None] 3.9939ms 3.6666ms 272.7320 Ops/s 265.5267 Ops/s $\color{#35bf28}+2.71\%$
test_redq_speed[True-backward] 9.1971ms 8.6545ms 115.5472 Ops/s 114.0435 Ops/s $\color{#35bf28}+1.32\%$
test_redq_speed[reduce-overhead-None] 3.9770ms 3.5386ms 282.5938 Ops/s 279.6662 Ops/s $\color{#35bf28}+1.05\%$
test_redq_speed[reduce-overhead-backward] 9.2860ms 8.7978ms 113.6644 Ops/s 114.8407 Ops/s $\color{#d91a1a}-1.02\%$
test_redq_deprec_speed[False-None] 11.0892ms 10.6814ms 93.6210 Ops/s 92.2399 Ops/s $\color{#35bf28}+1.50\%$
test_redq_deprec_speed[False-backward] 16.7716ms 15.7083ms 63.6606 Ops/s 64.0354 Ops/s $\color{#d91a1a}-0.59\%$
test_redq_deprec_speed[True-None] 3.6650ms 3.2664ms 306.1504 Ops/s 290.3256 Ops/s $\textbf{\color{#35bf28}+5.45\%}$
test_redq_deprec_speed[True-backward] 7.5691ms 7.1693ms 139.4831 Ops/s 131.9600 Ops/s $\textbf{\color{#35bf28}+5.70\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.5559ms 3.2138ms 311.1584 Ops/s 299.2747 Ops/s $\color{#35bf28}+3.97\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.5450ms 7.2073ms 138.7474 Ops/s 132.3718 Ops/s $\color{#35bf28}+4.82\%$
test_td3_speed[False-None] 7.6755ms 7.4649ms 133.9596 Ops/s 131.0657 Ops/s $\color{#35bf28}+2.21\%$
test_td3_speed[False-backward] 10.8784ms 10.3274ms 96.8294 Ops/s 94.5918 Ops/s $\color{#35bf28}+2.37\%$
test_td3_speed[True-None] 1.9581ms 1.9015ms 525.9109 Ops/s 523.2225 Ops/s $\color{#35bf28}+0.51\%$
test_td3_speed[True-backward] 4.0399ms 3.7476ms 266.8404 Ops/s 262.0108 Ops/s $\color{#35bf28}+1.84\%$
test_td3_speed[reduce-overhead-None] 1.9735ms 1.9100ms 523.5484 Ops/s 516.9515 Ops/s $\color{#35bf28}+1.28\%$
test_td3_speed[reduce-overhead-backward] 3.9456ms 3.7313ms 268.0055 Ops/s 263.0689 Ops/s $\color{#35bf28}+1.88\%$
test_cql_speed[False-None] 28.5028ms 25.4497ms 39.2932 Ops/s 39.6355 Ops/s $\color{#d91a1a}-0.86\%$
test_cql_speed[False-backward] 39.3090ms 35.4763ms 28.1878 Ops/s 28.5179 Ops/s $\color{#d91a1a}-1.16\%$
test_cql_speed[True-None] 11.5802ms 11.0970ms 90.1147 Ops/s 90.3501 Ops/s $\color{#d91a1a}-0.26\%$
test_cql_speed[True-backward] 17.5044ms 17.0290ms 58.7232 Ops/s 57.9411 Ops/s $\color{#35bf28}+1.35\%$
test_cql_speed[reduce-overhead-None] 12.1116ms 11.2835ms 88.6252 Ops/s 89.7896 Ops/s $\color{#d91a1a}-1.30\%$
test_cql_speed[reduce-overhead-backward] 17.6640ms 17.1310ms 58.3738 Ops/s 57.9490 Ops/s $\color{#35bf28}+0.73\%$
test_a2c_speed[False-None] 5.8261ms 5.3324ms 187.5344 Ops/s 187.4332 Ops/s $\color{#35bf28}+0.05\%$
test_a2c_speed[False-backward] 13.4403ms 11.8627ms 84.2977 Ops/s 84.2655 Ops/s $\color{#35bf28}+0.04\%$
test_a2c_speed[True-None] 3.4034ms 3.0809ms 324.5798 Ops/s 311.0078 Ops/s $\color{#35bf28}+4.36\%$
test_a2c_speed[True-backward] 8.9336ms 8.6425ms 115.7066 Ops/s 113.3844 Ops/s $\color{#35bf28}+2.05\%$
test_a2c_speed[reduce-overhead-None] 3.3734ms 3.1237ms 320.1307 Ops/s 321.2709 Ops/s $\color{#d91a1a}-0.35\%$
test_a2c_speed[reduce-overhead-backward] 8.9105ms 8.5167ms 117.4160 Ops/s 116.5541 Ops/s $\color{#35bf28}+0.74\%$
test_ppo_speed[False-None] 7.7287ms 5.6518ms 176.9361 Ops/s 172.6901 Ops/s $\color{#35bf28}+2.46\%$
test_ppo_speed[False-backward] 13.4930ms 12.4902ms 80.0625 Ops/s 80.0048 Ops/s $\color{#35bf28}+0.07\%$
test_ppo_speed[True-None] 3.7196ms 3.5446ms 282.1193 Ops/s 281.3224 Ops/s $\color{#35bf28}+0.28\%$
test_ppo_speed[True-backward] 9.0984ms 8.4768ms 117.9691 Ops/s 118.9118 Ops/s $\color{#d91a1a}-0.79\%$
test_ppo_speed[reduce-overhead-None] 3.6349ms 3.4719ms 288.0286 Ops/s 288.3710 Ops/s $\color{#d91a1a}-0.12\%$
test_ppo_speed[reduce-overhead-backward] 8.5656ms 8.3302ms 120.0453 Ops/s 116.7669 Ops/s $\color{#35bf28}+2.81\%$
test_reinforce_speed[False-None] 4.8164ms 4.4261ms 225.9319 Ops/s 216.2007 Ops/s $\color{#35bf28}+4.50\%$
test_reinforce_speed[False-backward] 7.9475ms 7.3939ms 135.2469 Ops/s 133.0445 Ops/s $\color{#35bf28}+1.66\%$
test_reinforce_speed[True-None] 2.6289ms 2.2310ms 448.2317 Ops/s 442.8853 Ops/s $\color{#35bf28}+1.21\%$
test_reinforce_speed[True-backward] 7.6531ms 7.2225ms 138.4558 Ops/s 138.0526 Ops/s $\color{#35bf28}+0.29\%$
test_reinforce_speed[reduce-overhead-None] 2.6218ms 2.2385ms 446.7342 Ops/s 439.7482 Ops/s $\color{#35bf28}+1.59\%$
test_reinforce_speed[reduce-overhead-backward] 7.5101ms 7.1633ms 139.6006 Ops/s 139.2302 Ops/s $\color{#35bf28}+0.27\%$
test_iql_speed[False-None] 25.0115ms 20.2390ms 49.4095 Ops/s 49.6314 Ops/s $\color{#d91a1a}-0.45\%$
test_iql_speed[False-backward] 35.8064ms 30.6196ms 32.6588 Ops/s 32.6061 Ops/s $\color{#35bf28}+0.16\%$
test_iql_speed[True-None] 7.1971ms 6.8281ms 146.4535 Ops/s 139.7402 Ops/s $\color{#35bf28}+4.80\%$
test_iql_speed[True-backward] 17.0743ms 15.8781ms 62.9796 Ops/s 61.7473 Ops/s $\color{#35bf28}+2.00\%$
test_iql_speed[reduce-overhead-None] 7.5349ms 6.8435ms 146.1239 Ops/s 146.1049 Ops/s $\color{#35bf28}+0.01\%$
test_iql_speed[reduce-overhead-backward] 16.4439ms 15.7916ms 63.3247 Ops/s 62.9217 Ops/s $\color{#35bf28}+0.64\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.4844ms 6.1407ms 162.8488 Ops/s 162.4107 Ops/s $\color{#35bf28}+0.27\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9781ms 0.3412ms 2.9304 KOps/s 3.5148 KOps/s $\textbf{\color{#d91a1a}-16.63\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6041ms 0.3204ms 3.1211 KOps/s 3.8431 KOps/s $\textbf{\color{#d91a1a}-18.79\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3766ms 5.9169ms 169.0079 Ops/s 170.6466 Ops/s $\color{#d91a1a}-0.96\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7759ms 0.3311ms 3.0204 KOps/s 3.4123 KOps/s $\textbf{\color{#d91a1a}-11.49\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6244ms 0.3153ms 3.1714 KOps/s 3.5435 KOps/s $\textbf{\color{#d91a1a}-10.50\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6305ms 1.3532ms 738.9879 Ops/s 750.5371 Ops/s $\color{#d91a1a}-1.54\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5839ms 1.3053ms 766.1256 Ops/s 784.3157 Ops/s $\color{#d91a1a}-2.32\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3704ms 6.1016ms 163.8911 Ops/s 166.0011 Ops/s $\color{#d91a1a}-1.27\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.2866ms 0.4263ms 2.3456 KOps/s 2.3043 KOps/s $\color{#35bf28}+1.80\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7774ms 0.3983ms 2.5104 KOps/s 2.3930 KOps/s $\color{#35bf28}+4.90\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2703ms 6.0123ms 166.3261 Ops/s 169.6240 Ops/s $\color{#d91a1a}-1.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.9453ms 0.3035ms 3.2947 KOps/s 3.7006 KOps/s $\textbf{\color{#d91a1a}-10.97\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5716ms 0.2860ms 3.4963 KOps/s 4.5677 KOps/s $\textbf{\color{#d91a1a}-23.46\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.1258ms 5.9084ms 169.2498 Ops/s 172.3526 Ops/s $\color{#d91a1a}-1.80\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.6317ms 0.3127ms 3.1979 KOps/s 3.0146 KOps/s $\textbf{\color{#35bf28}+6.08\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5288ms 0.2985ms 3.3498 KOps/s 3.3674 KOps/s $\color{#d91a1a}-0.52\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3394ms 6.0899ms 164.2071 Ops/s 167.3122 Ops/s $\color{#d91a1a}-1.86\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.6984ms 0.4217ms 2.3712 KOps/s 527.7710 Ops/s $\textbf{\color{#35bf28}+349.28\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 9.8143ms 0.4073ms 2.4551 KOps/s 2.2077 KOps/s $\textbf{\color{#35bf28}+11.21\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.9548ms 5.1744ms 193.2584 Ops/s 185.3580 Ops/s $\color{#35bf28}+4.26\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 6.9249ms 1.9458ms 513.9247 Ops/s 484.7695 Ops/s $\textbf{\color{#35bf28}+6.01\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.7875ms 1.2581ms 794.8525 Ops/s 845.6761 Ops/s $\textbf{\color{#d91a1a}-6.01\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.3957s 13.5874ms 73.5974 Ops/s 189.2881 Ops/s $\textbf{\color{#d91a1a}-61.12\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.0584ms 2.0044ms 498.8925 Ops/s 509.3968 Ops/s $\color{#d91a1a}-2.06\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.8346ms 1.2012ms 832.5160 Ops/s 789.6814 Ops/s $\textbf{\color{#35bf28}+5.42\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.3166ms 5.4196ms 184.5148 Ops/s 181.6590 Ops/s $\color{#35bf28}+1.57\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.9592ms 2.1335ms 468.7143 Ops/s 457.3391 Ops/s $\color{#35bf28}+2.49\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.0960ms 1.3697ms 730.0852 Ops/s 721.9556 Ops/s $\color{#35bf28}+1.13\%$

@vmoens vmoens merged commit baba52b into main Oct 21, 2024
70 of 75 checks passed
@vmoens vmoens deleted the fix-benchmarks branch October 21, 2024 14:23
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

点击 这是indexloc提供的php浏览器服务,不要输入任何密码和下载