Skip to content

Conversation

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Dec 31, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3275

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

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

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 31, 2025
@vmoens vmoens merged commit 9516ea8 into gh/vmoens/175/base Dec 31, 2025
78 of 93 checks passed
@vmoens vmoens deleted the gh/vmoens/175/head branch December 31, 2025 09:17
@github-actions
Copy link

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

Total Benchmarks: 164. Improved: $\large\color{#35bf28}23$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 80.2416μs 79.4343μs 12.5890 KOps/s 12.5683 KOps/s $\color{#35bf28}+0.17\%$
test_tensor_to_bytestream_speed[torch.save] 0.1389ms 0.1383ms 7.2312 KOps/s 7.1273 KOps/s $\color{#35bf28}+1.46\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1178s 0.1174s 8.5181 Ops/s 8.8743 Ops/s $\color{#d91a1a}-4.01\%$
test_tensor_to_bytestream_speed[numpy] 2.8031μs 2.7974μs 357.4719 KOps/s 385.1187 KOps/s $\textbf{\color{#d91a1a}-7.18\%}$
test_tensor_to_bytestream_speed[safetensors] 39.2196μs 39.0432μs 25.6126 KOps/s 26.6515 KOps/s $\color{#d91a1a}-3.90\%$
test_simple 0.5412s 0.5387s 1.8565 Ops/s 1.7899 Ops/s $\color{#35bf28}+3.72\%$
test_transformed 1.1101s 1.1076s 0.9028 Ops/s 0.8930 Ops/s $\color{#35bf28}+1.10\%$
test_serial 1.6408s 1.6373s 0.6108 Ops/s 0.6089 Ops/s $\color{#35bf28}+0.30\%$
test_parallel 1.3810s 1.1866s 0.8428 Ops/s 0.8903 Ops/s $\textbf{\color{#d91a1a}-5.34\%}$
test_step_mdp_speed[True-True-True-True-True] 0.1812ms 43.1971μs 23.1497 KOps/s 22.8444 KOps/s $\color{#35bf28}+1.34\%$
test_step_mdp_speed[True-True-True-True-False] 60.5910μs 24.7600μs 40.3877 KOps/s 40.3878 KOps/s $-0.00\%$
test_step_mdp_speed[True-True-True-False-True] 61.5610μs 23.9226μs 41.8015 KOps/s 40.6810 KOps/s $\color{#35bf28}+2.75\%$
test_step_mdp_speed[True-True-True-False-False] 43.4600μs 13.3107μs 75.1277 KOps/s 75.1082 KOps/s $\color{#35bf28}+0.03\%$
test_step_mdp_speed[True-True-False-True-True] 89.8520μs 46.1322μs 21.6768 KOps/s 21.5681 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[True-True-False-True-False] 60.8010μs 27.1787μs 36.7936 KOps/s 37.4934 KOps/s $\color{#d91a1a}-1.87\%$
test_step_mdp_speed[True-True-False-False-True] 68.4910μs 27.5110μs 36.3492 KOps/s 37.0577 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[True-True-False-False-False] 40.7200μs 16.1201μs 62.0345 KOps/s 61.5270 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-False-True-True-True] 92.2010μs 49.3461μs 20.2650 KOps/s 20.0435 KOps/s $\color{#35bf28}+1.11\%$
test_step_mdp_speed[True-False-True-True-False] 63.3420μs 29.8208μs 33.5337 KOps/s 33.5706 KOps/s $\color{#d91a1a}-0.11\%$
test_step_mdp_speed[True-False-True-False-True] 67.2210μs 27.3431μs 36.5722 KOps/s 36.7366 KOps/s $\color{#d91a1a}-0.45\%$
test_step_mdp_speed[True-False-True-False-False] 48.0200μs 16.0618μs 62.2594 KOps/s 62.0353 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[True-False-False-True-True] 90.4610μs 52.0174μs 19.2243 KOps/s 19.2355 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[True-False-False-True-False] 60.1720μs 32.3992μs 30.8650 KOps/s 31.0133 KOps/s $\color{#d91a1a}-0.48\%$
test_step_mdp_speed[True-False-False-False-True] 61.6920μs 29.9021μs 33.4424 KOps/s 33.9151 KOps/s $\color{#d91a1a}-1.39\%$
test_step_mdp_speed[True-False-False-False-False] 46.4710μs 18.8029μs 53.1831 KOps/s 52.6814 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[False-True-True-True-True] 91.9720μs 49.0442μs 20.3898 KOps/s 20.0037 KOps/s $\color{#35bf28}+1.93\%$
test_step_mdp_speed[False-True-True-True-False] 63.8310μs 29.7271μs 33.6393 KOps/s 33.5931 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[False-True-True-False-True] 2.4027ms 31.2693μs 31.9803 KOps/s 32.1729 KOps/s $\color{#d91a1a}-0.60\%$
test_step_mdp_speed[False-True-True-False-False] 54.6010μs 17.8859μs 55.9101 KOps/s 55.9391 KOps/s $\color{#d91a1a}-0.05\%$
test_step_mdp_speed[False-True-False-True-True] 86.7420μs 53.0063μs 18.8657 KOps/s 18.9669 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[False-True-False-True-False] 70.2410μs 33.2232μs 30.0994 KOps/s 30.8533 KOps/s $\color{#d91a1a}-2.44\%$
test_step_mdp_speed[False-True-False-False-True] 74.7820μs 33.4272μs 29.9157 KOps/s 29.4875 KOps/s $\color{#35bf28}+1.45\%$
test_step_mdp_speed[False-True-False-False-False] 54.0010μs 20.8470μs 47.9686 KOps/s 48.7314 KOps/s $\color{#d91a1a}-1.57\%$
test_step_mdp_speed[False-False-True-True-True] 90.0320μs 55.1584μs 18.1296 KOps/s 18.2602 KOps/s $\color{#d91a1a}-0.72\%$
test_step_mdp_speed[False-False-True-True-False] 64.6720μs 35.6735μs 28.0320 KOps/s 28.1804 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[False-False-True-False-True] 65.3120μs 33.3334μs 29.9999 KOps/s 29.9134 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[False-False-True-False-False] 57.0910μs 20.4169μs 48.9790 KOps/s 49.0381 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-False-False-True-True] 0.1139ms 57.0559μs 17.5267 KOps/s 17.6077 KOps/s $\color{#d91a1a}-0.46\%$
test_step_mdp_speed[False-False-False-True-False] 75.1010μs 37.6970μs 26.5273 KOps/s 26.4145 KOps/s $\color{#35bf28}+0.43\%$
test_step_mdp_speed[False-False-False-False-True] 74.1220μs 35.4318μs 28.2232 KOps/s 28.2905 KOps/s $\color{#d91a1a}-0.24\%$
test_step_mdp_speed[False-False-False-False-False] 95.8420μs 21.7468μs 45.9838 KOps/s 42.8796 KOps/s $\textbf{\color{#35bf28}+7.24\%}$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8653s 0.7693s 1.2999 Ops/s 1.3067 Ops/s $\color{#d91a1a}-0.52\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7232s 0.6399s 1.5626 Ops/s 1.5824 Ops/s $\color{#d91a1a}-1.25\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.8053s 1.7086s 0.5853 Ops/s 0.6005 Ops/s $\color{#d91a1a}-2.53\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5305s 1.4534s 0.6880 Ops/s 0.6904 Ops/s $\color{#d91a1a}-0.34\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0122s 1.9261s 0.5192 Ops/s 0.5264 Ops/s $\color{#d91a1a}-1.37\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7783s 1.6924s 0.5909 Ops/s 0.5915 Ops/s $\color{#d91a1a}-0.11\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.7443s 4.6395s 0.2155 Ops/s 0.2155 Ops/s $\color{#35bf28}+0.02\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.5519s 4.4062s 0.2270 Ops/s 0.2231 Ops/s $\color{#35bf28}+1.72\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0163s 1.9274s 0.5188 Ops/s 0.5208 Ops/s $\color{#d91a1a}-0.38\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7450s 1.6657s 0.6004 Ops/s 0.6131 Ops/s $\color{#d91a1a}-2.08\%$
test_values[generalized_advantage_estimate-True-True] 10.5349ms 10.4086ms 96.0747 Ops/s 100.9084 Ops/s $\color{#d91a1a}-4.79\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.9887ms 18.1156ms 55.2009 Ops/s 56.5429 Ops/s $\color{#d91a1a}-2.37\%$
test_values[td0_return_estimate-False-False] 0.2272ms 0.1314ms 7.6076 KOps/s 7.7388 KOps/s $\color{#d91a1a}-1.69\%$
test_values[td1_return_estimate-False-False] 27.7672ms 27.3403ms 36.5761 Ops/s 36.9426 Ops/s $\color{#d91a1a}-0.99\%$
test_values[vec_td1_return_estimate-False-False] 18.5605ms 17.7552ms 56.3216 Ops/s 56.6180 Ops/s $\color{#d91a1a}-0.52\%$
test_values[td_lambda_return_estimate-True-False] 41.3477ms 40.3909ms 24.7581 Ops/s 24.9930 Ops/s $\color{#d91a1a}-0.94\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.2131ms 17.8464ms 56.0336 Ops/s 56.4761 Ops/s $\color{#d91a1a}-0.78\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.4650ms 9.2013ms 108.6802 Ops/s 113.2416 Ops/s $\color{#d91a1a}-4.03\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7080ms 1.4716ms 679.5124 Ops/s 675.6169 Ops/s $\color{#35bf28}+0.58\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5127ms 0.4220ms 2.3698 KOps/s 2.4665 KOps/s $\color{#d91a1a}-3.92\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.3091ms 32.1244ms 31.1290 Ops/s 28.8319 Ops/s $\textbf{\color{#35bf28}+7.97\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9526ms 1.7245ms 579.8690 Ops/s 553.4757 Ops/s $\color{#35bf28}+4.77\%$
test_dqn_speed[False-None] 1.6470ms 1.4043ms 712.0829 Ops/s 714.9307 Ops/s $\color{#d91a1a}-0.40\%$
test_dqn_speed[False-backward] 2.0399ms 1.9243ms 519.6775 Ops/s 528.0813 Ops/s $\color{#d91a1a}-1.59\%$
test_dqn_speed[True-None] 0.7401ms 0.5245ms 1.9067 KOps/s 1.8237 KOps/s $\color{#35bf28}+4.55\%$
test_dqn_speed[True-backward] 1.0197ms 0.9661ms 1.0351 KOps/s 896.2484 Ops/s $\textbf{\color{#35bf28}+15.49\%}$
test_dqn_speed[reduce-overhead-None] 0.6289ms 0.5098ms 1.9617 KOps/s 1.9031 KOps/s $\color{#35bf28}+3.08\%$
test_dqn_speed[reduce-overhead-backward] 0.9928ms 0.9433ms 1.0601 KOps/s 937.5686 Ops/s $\textbf{\color{#35bf28}+13.07\%}$
test_ddpg_speed[False-None] 3.2335ms 2.8399ms 352.1200 Ops/s 347.5834 Ops/s $\color{#35bf28}+1.31\%$
test_ddpg_speed[False-backward] 4.3563ms 4.1149ms 243.0165 Ops/s 247.3042 Ops/s $\color{#d91a1a}-1.73\%$
test_ddpg_speed[True-None] 1.4592ms 1.3518ms 739.7295 Ops/s 726.5070 Ops/s $\color{#35bf28}+1.82\%$
test_ddpg_speed[True-backward] 2.4515ms 2.3319ms 428.8304 Ops/s 353.2839 Ops/s $\textbf{\color{#35bf28}+21.38\%}$
test_ddpg_speed[reduce-overhead-None] 1.4871ms 1.3490ms 741.2645 Ops/s 738.5984 Ops/s $\color{#35bf28}+0.36\%$
test_ddpg_speed[reduce-overhead-backward] 2.4695ms 2.3195ms 431.1280 Ops/s 353.3771 Ops/s $\textbf{\color{#35bf28}+22.00\%}$
test_sac_speed[False-None] 8.3106ms 7.8149ms 127.9612 Ops/s 124.2237 Ops/s $\color{#35bf28}+3.01\%$
test_sac_speed[False-backward] 11.5571ms 11.0692ms 90.3407 Ops/s 90.6196 Ops/s $\color{#d91a1a}-0.31\%$
test_sac_speed[True-None] 2.2644ms 2.1012ms 475.9234 Ops/s 466.5655 Ops/s $\color{#35bf28}+2.01\%$
test_sac_speed[True-backward] 4.1053ms 3.9713ms 251.8071 Ops/s 250.2167 Ops/s $\color{#35bf28}+0.64\%$
test_sac_speed[reduce-overhead-None] 2.2318ms 2.0994ms 476.3368 Ops/s 459.5696 Ops/s $\color{#35bf28}+3.65\%$
test_sac_speed[reduce-overhead-backward] 4.1200ms 3.9880ms 250.7495 Ops/s 203.3842 Ops/s $\textbf{\color{#35bf28}+23.29\%}$
test_redq_speed[False-None] 13.5792ms 10.3857ms 96.2860 Ops/s 98.4435 Ops/s $\color{#d91a1a}-2.19\%$
test_redq_speed[False-backward] 21.8652ms 17.9218ms 55.7978 Ops/s 56.8488 Ops/s $\color{#d91a1a}-1.85\%$
test_redq_speed[True-None] 4.6000ms 4.3577ms 229.4788 Ops/s 213.5150 Ops/s $\textbf{\color{#35bf28}+7.48\%}$
test_redq_speed[True-backward] 10.3510ms 9.4698ms 105.5993 Ops/s 98.9482 Ops/s $\textbf{\color{#35bf28}+6.72\%}$
test_redq_speed[reduce-overhead-None] 4.5180ms 4.2809ms 233.5954 Ops/s 212.6223 Ops/s $\textbf{\color{#35bf28}+9.86\%}$
test_redq_speed[reduce-overhead-backward] 9.9301ms 9.6220ms 103.9280 Ops/s 102.6472 Ops/s $\color{#35bf28}+1.25\%$
test_redq_deprec_speed[False-None] 11.4561ms 10.7314ms 93.1848 Ops/s 92.1494 Ops/s $\color{#35bf28}+1.12\%$
test_redq_deprec_speed[False-backward] 16.2317ms 15.4195ms 64.8529 Ops/s 64.2990 Ops/s $\color{#35bf28}+0.86\%$
test_redq_deprec_speed[True-None] 3.7532ms 3.5427ms 282.2667 Ops/s 287.8602 Ops/s $\color{#d91a1a}-1.94\%$
test_redq_deprec_speed[True-backward] 7.6916ms 7.3658ms 135.7632 Ops/s 138.2275 Ops/s $\color{#d91a1a}-1.78\%$
test_redq_deprec_speed[reduce-overhead-None] 3.6604ms 3.4626ms 288.7978 Ops/s 283.3608 Ops/s $\color{#35bf28}+1.92\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.6419ms 7.3504ms 136.0476 Ops/s 125.0908 Ops/s $\textbf{\color{#35bf28}+8.76\%}$
test_td3_speed[False-None] 8.1359ms 7.9212ms 126.2441 Ops/s 125.9217 Ops/s $\color{#35bf28}+0.26\%$
test_td3_speed[False-backward] 11.2691ms 10.8230ms 92.3954 Ops/s 93.1006 Ops/s $\color{#d91a1a}-0.76\%$
test_td3_speed[True-None] 1.8290ms 1.7810ms 561.4760 Ops/s 555.9120 Ops/s $\color{#35bf28}+1.00\%$
test_td3_speed[True-backward] 3.9025ms 3.5715ms 279.9964 Ops/s 277.3623 Ops/s $\color{#35bf28}+0.95\%$
test_td3_speed[reduce-overhead-None] 1.8470ms 1.7707ms 564.7512 Ops/s 567.8299 Ops/s $\color{#d91a1a}-0.54\%$
test_td3_speed[reduce-overhead-backward] 3.9939ms 3.6485ms 274.0828 Ops/s 260.4693 Ops/s $\textbf{\color{#35bf28}+5.23\%}$
test_cql_speed[False-None] 29.1819ms 25.8294ms 38.7155 Ops/s 39.2365 Ops/s $\color{#d91a1a}-1.33\%$
test_cql_speed[False-backward] 35.6598ms 34.7256ms 28.7972 Ops/s 28.8720 Ops/s $\color{#d91a1a}-0.26\%$
test_cql_speed[True-None] 12.5908ms 12.2575ms 81.5825 Ops/s 79.4416 Ops/s $\color{#35bf28}+2.69\%$
test_cql_speed[True-backward] 18.5338ms 17.9912ms 55.5827 Ops/s 53.4901 Ops/s $\color{#35bf28}+3.91\%$
test_cql_speed[reduce-overhead-None] 12.7277ms 12.3349ms 81.0705 Ops/s 80.0158 Ops/s $\color{#35bf28}+1.32\%$
test_cql_speed[reduce-overhead-backward] 18.6367ms 17.9696ms 55.6494 Ops/s 55.4756 Ops/s $\color{#35bf28}+0.31\%$
test_a2c_speed[False-None] 5.6211ms 5.3421ms 187.1938 Ops/s 185.7170 Ops/s $\color{#35bf28}+0.80\%$
test_a2c_speed[False-backward] 12.2601ms 11.8164ms 84.6284 Ops/s 84.9720 Ops/s $\color{#d91a1a}-0.40\%$
test_a2c_speed[True-None] 3.7556ms 3.6074ms 277.2112 Ops/s 260.5584 Ops/s $\textbf{\color{#35bf28}+6.39\%}$
test_a2c_speed[True-backward] 8.7538ms 8.4761ms 117.9793 Ops/s 109.2070 Ops/s $\textbf{\color{#35bf28}+8.03\%}$
test_a2c_speed[reduce-overhead-None] 3.8191ms 3.6514ms 273.8661 Ops/s 274.8860 Ops/s $\color{#d91a1a}-0.37\%$
test_a2c_speed[reduce-overhead-backward] 8.9354ms 8.6036ms 116.2304 Ops/s 115.2021 Ops/s $\color{#35bf28}+0.89\%$
test_ppo_speed[False-None] 5.9544ms 5.7415ms 174.1719 Ops/s 171.2207 Ops/s $\color{#35bf28}+1.72\%$
test_ppo_speed[False-backward] 12.6009ms 12.3303ms 81.1010 Ops/s 79.9083 Ops/s $\color{#35bf28}+1.49\%$
test_ppo_speed[True-None] 3.8297ms 3.5606ms 280.8527 Ops/s 271.1508 Ops/s $\color{#35bf28}+3.58\%$
test_ppo_speed[True-backward] 8.6050ms 8.3523ms 119.7275 Ops/s 112.4288 Ops/s $\textbf{\color{#35bf28}+6.49\%}$
test_ppo_speed[reduce-overhead-None] 3.9051ms 3.5326ms 283.0788 Ops/s 282.4105 Ops/s $\color{#35bf28}+0.24\%$
test_ppo_speed[reduce-overhead-backward] 8.8321ms 8.5993ms 116.2891 Ops/s 108.8381 Ops/s $\textbf{\color{#35bf28}+6.85\%}$
test_reinforce_speed[False-None] 4.6770ms 4.4748ms 223.4757 Ops/s 222.6161 Ops/s $\color{#35bf28}+0.39\%$
test_reinforce_speed[False-backward] 7.5350ms 7.2797ms 137.3680 Ops/s 136.3177 Ops/s $\color{#35bf28}+0.77\%$
test_reinforce_speed[True-None] 3.0630ms 2.8101ms 355.8573 Ops/s 352.1921 Ops/s $\color{#35bf28}+1.04\%$
test_reinforce_speed[True-backward] 7.8215ms 7.5553ms 132.3567 Ops/s 132.1655 Ops/s $\color{#35bf28}+0.14\%$
test_reinforce_speed[reduce-overhead-None] 2.9593ms 2.7879ms 358.6993 Ops/s 356.0847 Ops/s $\color{#35bf28}+0.73\%$
test_reinforce_speed[reduce-overhead-backward] 8.1181ms 7.8257ms 127.7847 Ops/s 124.5490 Ops/s $\color{#35bf28}+2.60\%$
test_iql_speed[False-None] 20.0840ms 19.4793ms 51.3364 Ops/s 49.6013 Ops/s $\color{#35bf28}+3.50\%$
test_iql_speed[False-backward] 30.7115ms 30.1212ms 33.1992 Ops/s 32.8243 Ops/s $\color{#35bf28}+1.14\%$
test_iql_speed[True-None] 9.9392ms 8.4864ms 117.8354 Ops/s 114.2358 Ops/s $\color{#35bf28}+3.15\%$
test_iql_speed[True-backward] 16.7278ms 16.3624ms 61.1156 Ops/s 60.0319 Ops/s $\color{#35bf28}+1.81\%$
test_iql_speed[reduce-overhead-None] 8.6814ms 8.4342ms 118.5655 Ops/s 115.3897 Ops/s $\color{#35bf28}+2.75\%$
test_iql_speed[reduce-overhead-backward] 17.0491ms 16.7337ms 59.7595 Ops/s 56.8863 Ops/s $\textbf{\color{#35bf28}+5.05\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9897ms 6.0417ms 165.5156 Ops/s 167.0768 Ops/s $\color{#d91a1a}-0.93\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6183ms 0.3644ms 2.7442 KOps/s 2.9182 KOps/s $\textbf{\color{#d91a1a}-5.96\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6338ms 0.3493ms 2.8628 KOps/s 3.0686 KOps/s $\textbf{\color{#d91a1a}-6.70\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9473ms 5.6785ms 176.1029 Ops/s 174.3470 Ops/s $\color{#35bf28}+1.01\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0998ms 0.3557ms 2.8111 KOps/s 2.8222 KOps/s $\color{#d91a1a}-0.39\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5693ms 0.3429ms 2.9166 KOps/s 2.9681 KOps/s $\color{#d91a1a}-1.74\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6717ms 1.4120ms 708.2388 Ops/s 719.6971 Ops/s $\color{#d91a1a}-1.59\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6569ms 1.3369ms 748.0044 Ops/s 767.7975 Ops/s $\color{#d91a1a}-2.58\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.7108ms 6.0262ms 165.9417 Ops/s 168.6294 Ops/s $\color{#d91a1a}-1.59\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8566ms 0.4680ms 2.1366 KOps/s 1.9654 KOps/s $\textbf{\color{#35bf28}+8.71\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7007ms 0.4187ms 2.3885 KOps/s 2.0440 KOps/s $\textbf{\color{#35bf28}+16.85\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.8256ms 5.7225ms 174.7492 Ops/s 172.3831 Ops/s $\color{#35bf28}+1.37\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0117ms 0.2910ms 3.4361 KOps/s 2.7295 KOps/s $\textbf{\color{#35bf28}+25.89\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5593ms 0.3178ms 3.1469 KOps/s 2.8703 KOps/s $\textbf{\color{#35bf28}+9.64\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0973ms 5.7055ms 175.2692 Ops/s 174.2304 Ops/s $\color{#35bf28}+0.60\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.6578ms 0.3031ms 3.2989 KOps/s 3.5910 KOps/s $\textbf{\color{#d91a1a}-8.13\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5717ms 0.3454ms 2.8948 KOps/s 3.2492 KOps/s $\textbf{\color{#d91a1a}-10.91\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0818ms 5.9349ms 168.4941 Ops/s 169.9834 Ops/s $\color{#d91a1a}-0.88\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.3699ms 0.5129ms 1.9496 KOps/s 2.0956 KOps/s $\textbf{\color{#d91a1a}-6.97\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7097ms 0.4971ms 2.0117 KOps/s 2.2142 KOps/s $\textbf{\color{#d91a1a}-9.15\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4101ms 4.9451ms 202.2219 Ops/s 195.4431 Ops/s $\color{#35bf28}+3.47\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.3815ms 2.3231ms 430.4663 Ops/s 430.0661 Ops/s $\color{#35bf28}+0.09\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.0364ms 1.2189ms 820.3850 Ops/s 855.0437 Ops/s $\color{#d91a1a}-4.05\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 7.0290ms 5.0432ms 198.2879 Ops/s 51.3101 Ops/s $\textbf{\color{#35bf28}+286.45\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.6919ms 2.3575ms 424.1706 Ops/s 582.3083 Ops/s $\textbf{\color{#d91a1a}-27.16\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.3103ms 1.2148ms 823.1650 Ops/s 981.6857 Ops/s $\textbf{\color{#d91a1a}-16.15\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.6226s 17.6009ms 56.8152 Ops/s 189.3795 Ops/s $\textbf{\color{#d91a1a}-70.00\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 12.0678ms 2.0992ms 476.3607 Ops/s 457.3400 Ops/s $\color{#35bf28}+4.16\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 1.3506ms 1.0482ms 954.0191 Ops/s 727.6460 Ops/s $\textbf{\color{#35bf28}+31.11\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 36.5621ms 34.1119ms 29.3153 Ops/s 29.2943 Ops/s $\color{#35bf28}+0.07\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.4515ms 17.6335ms 56.7102 Ops/s 56.0664 Ops/s $\color{#35bf28}+1.15\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 36.5710ms 34.5979ms 28.9035 Ops/s 28.1539 Ops/s $\color{#35bf28}+2.66\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.3793ms 17.7440ms 56.3570 Ops/s 55.3230 Ops/s $\color{#35bf28}+1.87\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 37.9265ms 36.3327ms 27.5234 Ops/s 26.4207 Ops/s $\color{#35bf28}+4.17\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.1690ms 19.0316ms 52.5442 Ops/s 51.8605 Ops/s $\color{#35bf28}+1.32\%$

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.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants