Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature] DataLoadingPrimer handling of dataloader with batch-size > 0 #2821

Merged
merged 2 commits into from
Mar 3, 2025

Conversation

[ghstack-poisoned]
Copy link

pytorch-bot bot commented Mar 3, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2821

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

❌ 3 New Failures, 8 Pending

As of commit 1541dd5 with merge base 8c9dc05 (image):

NEW FAILURES - The following jobs have failed:

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 Mar 3, 2025
vmoens added a commit that referenced this pull request Mar 3, 2025
ghstack-source-id: f288a544c0dd2e8d4d5e1fed70e1e28d70c22aea
Pull Request resolved: #2821
Copy link

github-actions bot commented Mar 3, 2025

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}29$. Worsened: $\large\color{#d91a1a}2$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.6460s 0.5358s 1.8663 Ops/s 1.8945 Ops/s $\color{#d91a1a}-1.49\%$
test_transformed 1.1342s 1.0259s 0.9748 Ops/s 0.9775 Ops/s $\color{#d91a1a}-0.28\%$
test_serial 1.5222s 1.5086s 0.6629 Ops/s 0.6648 Ops/s $\color{#d91a1a}-0.29\%$
test_parallel 1.4506s 1.3403s 0.7461 Ops/s 0.7486 Ops/s $\color{#d91a1a}-0.34\%$
test_step_mdp_speed[True-True-True-True-True] 0.1649ms 29.9062μs 33.4379 KOps/s 33.5429 KOps/s $\color{#d91a1a}-0.31\%$
test_step_mdp_speed[True-True-True-True-False] 79.7200μs 17.9224μs 55.7961 KOps/s 57.2521 KOps/s $\color{#d91a1a}-2.54\%$
test_step_mdp_speed[True-True-True-False-True] 49.3130μs 16.9133μs 59.1252 KOps/s 59.5101 KOps/s $\color{#d91a1a}-0.65\%$
test_step_mdp_speed[True-True-True-False-False] 84.4290μs 10.0350μs 99.6513 KOps/s 101.1316 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[True-True-False-True-True] 88.3260μs 32.7106μs 30.5712 KOps/s 31.6087 KOps/s $\color{#d91a1a}-3.28\%$
test_step_mdp_speed[True-True-False-True-False] 81.7240μs 19.8480μs 50.3829 KOps/s 51.6405 KOps/s $\color{#d91a1a}-2.44\%$
test_step_mdp_speed[True-True-False-False-True] 95.3090μs 18.9285μs 52.8305 KOps/s 53.2275 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[True-True-False-False-False] 59.4010μs 11.9576μs 83.6292 KOps/s 84.7855 KOps/s $\color{#d91a1a}-1.36\%$
test_step_mdp_speed[True-False-True-True-True] 99.3670μs 34.0660μs 29.3548 KOps/s 29.6278 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[True-False-True-True-False] 53.0700μs 21.8245μs 45.8201 KOps/s 46.6893 KOps/s $\color{#d91a1a}-1.86\%$
test_step_mdp_speed[True-False-True-False-True] 0.1009ms 18.8811μs 52.9629 KOps/s 53.4466 KOps/s $\color{#d91a1a}-0.90\%$
test_step_mdp_speed[True-False-True-False-False] 40.9170μs 12.1140μs 82.5493 KOps/s 84.9220 KOps/s $\color{#d91a1a}-2.79\%$
test_step_mdp_speed[True-False-False-True-True] 0.1183ms 35.4067μs 28.2433 KOps/s 28.2966 KOps/s $\color{#d91a1a}-0.19\%$
test_step_mdp_speed[True-False-False-True-False] 80.9320μs 23.6223μs 42.3329 KOps/s 43.3686 KOps/s $\color{#d91a1a}-2.39\%$
test_step_mdp_speed[True-False-False-False-True] 91.3530μs 20.6460μs 48.4355 KOps/s 48.6786 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-False-False-False-False] 71.7950μs 13.6802μs 73.0985 KOps/s 74.2801 KOps/s $\color{#d91a1a}-1.59\%$
test_step_mdp_speed[False-True-True-True-True] 78.7180μs 34.1538μs 29.2793 KOps/s 29.6466 KOps/s $\color{#d91a1a}-1.24\%$
test_step_mdp_speed[False-True-True-True-False] 79.7100μs 21.8930μs 45.6768 KOps/s 46.6580 KOps/s $\color{#d91a1a}-2.10\%$
test_step_mdp_speed[False-True-True-False-True] 68.8400μs 21.7404μs 45.9972 KOps/s 46.3726 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[False-True-True-False-False] 73.7790μs 13.2894μs 75.2480 KOps/s 76.5728 KOps/s $\color{#d91a1a}-1.73\%$
test_step_mdp_speed[False-True-False-True-True] 78.0870μs 35.8704μs 27.8781 KOps/s 28.2985 KOps/s $\color{#d91a1a}-1.49\%$
test_step_mdp_speed[False-True-False-True-False] 95.6000μs 23.6370μs 42.3066 KOps/s 43.2065 KOps/s $\color{#d91a1a}-2.08\%$
test_step_mdp_speed[False-True-False-False-True] 2.7405ms 23.4489μs 42.6460 KOps/s 43.7136 KOps/s $\color{#d91a1a}-2.44\%$
test_step_mdp_speed[False-True-False-False-False] 49.3130μs 15.4209μs 64.8470 KOps/s 67.2918 KOps/s $\color{#d91a1a}-3.63\%$
test_step_mdp_speed[False-False-True-True-True] 0.1036ms 37.7476μs 26.4917 KOps/s 26.8397 KOps/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[False-False-True-True-False] 76.1230μs 25.9937μs 38.4709 KOps/s 39.8532 KOps/s $\color{#d91a1a}-3.47\%$
test_step_mdp_speed[False-False-True-False-True] 90.5710μs 23.3122μs 42.8960 KOps/s 43.5116 KOps/s $\color{#d91a1a}-1.41\%$
test_step_mdp_speed[False-False-True-False-False] 54.6330μs 15.4208μs 64.8473 KOps/s 67.0176 KOps/s $\color{#d91a1a}-3.24\%$
test_step_mdp_speed[False-False-False-True-True] 0.1128ms 39.4888μs 25.3236 KOps/s 25.1253 KOps/s $\color{#35bf28}+0.79\%$
test_step_mdp_speed[False-False-False-True-False] 70.8140μs 27.2591μs 36.6850 KOps/s 35.1860 KOps/s $\color{#35bf28}+4.26\%$
test_step_mdp_speed[False-False-False-False-True] 91.4220μs 24.7581μs 40.3908 KOps/s 41.0737 KOps/s $\color{#d91a1a}-1.66\%$
test_step_mdp_speed[False-False-False-False-False] 46.5680μs 16.7843μs 59.5795 KOps/s 61.0581 KOps/s $\color{#d91a1a}-2.42\%$
test_values[generalized_advantage_estimate-True-True] 10.7569ms 10.0489ms 99.5131 Ops/s 99.0823 Ops/s $\color{#35bf28}+0.43\%$
test_values[vec_generalized_advantage_estimate-True-True] 27.8510ms 24.8680ms 40.2124 Ops/s 37.4111 Ops/s $\textbf{\color{#35bf28}+7.49\%}$
test_values[td0_return_estimate-False-False] 0.2940ms 0.2005ms 4.9887 KOps/s 5.1198 KOps/s $\color{#d91a1a}-2.56\%$
test_values[td1_return_estimate-False-False] 26.3490ms 24.6798ms 40.5190 Ops/s 40.1030 Ops/s $\color{#35bf28}+1.04\%$
test_values[vec_td1_return_estimate-False-False] 26.2430ms 24.8508ms 40.2402 Ops/s 36.8648 Ops/s $\textbf{\color{#35bf28}+9.16\%}$
test_values[td_lambda_return_estimate-True-False] 36.3572ms 35.1008ms 28.4894 Ops/s 28.3168 Ops/s $\color{#35bf28}+0.61\%$
test_values[vec_td_lambda_return_estimate-True-False] 27.8024ms 24.9016ms 40.1581 Ops/s 37.4250 Ops/s $\textbf{\color{#35bf28}+7.30\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.9999ms 8.5775ms 116.5847 Ops/s 114.8182 Ops/s $\color{#35bf28}+1.54\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4067ms 1.7432ms 573.6664 Ops/s 542.2026 Ops/s $\textbf{\color{#35bf28}+5.80\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.5159ms 0.3697ms 2.7047 KOps/s 2.7070 KOps/s $\color{#d91a1a}-0.08\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 44.1061ms 40.0574ms 24.9642 Ops/s 22.5893 Ops/s $\textbf{\color{#35bf28}+10.51\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.6850ms 3.6108ms 276.9447 Ops/s 281.3384 Ops/s $\color{#d91a1a}-1.56\%$
test_dqn_speed[False-None] 1.9629ms 1.4239ms 702.2885 Ops/s 710.1292 Ops/s $\color{#d91a1a}-1.10\%$
test_dqn_speed[False-backward] 1.9721ms 1.9124ms 522.9137 Ops/s 529.1352 Ops/s $\color{#d91a1a}-1.18\%$
test_dqn_speed[True-None] 0.7557ms 0.4799ms 2.0837 KOps/s 2.0258 KOps/s $\color{#35bf28}+2.86\%$
test_dqn_speed[True-backward] 1.1578ms 0.9551ms 1.0470 KOps/s 1.0912 KOps/s $\color{#d91a1a}-4.05\%$
test_dqn_speed[reduce-overhead-None] 0.7313ms 0.4855ms 2.0596 KOps/s 2.0625 KOps/s $\color{#d91a1a}-0.14\%$
test_dqn_speed[reduce-overhead-backward] 1.0654ms 0.9246ms 1.0815 KOps/s 1.0075 KOps/s $\textbf{\color{#35bf28}+7.35\%}$
test_ddpg_speed[False-None] 3.4053ms 2.9605ms 337.7793 Ops/s 336.3500 Ops/s $\color{#35bf28}+0.42\%$
test_ddpg_speed[False-backward] 5.3050ms 4.2352ms 236.1184 Ops/s 241.3378 Ops/s $\color{#d91a1a}-2.16\%$
test_ddpg_speed[True-None] 1.4082ms 1.2393ms 806.9199 Ops/s 790.6051 Ops/s $\color{#35bf28}+2.06\%$
test_ddpg_speed[True-backward] 2.3891ms 2.2061ms 453.2873 Ops/s 448.2229 Ops/s $\color{#35bf28}+1.13\%$
test_ddpg_speed[reduce-overhead-None] 1.7405ms 1.2457ms 802.7318 Ops/s 797.0864 Ops/s $\color{#35bf28}+0.71\%$
test_ddpg_speed[reduce-overhead-backward] 3.3637ms 2.1907ms 456.4743 Ops/s 449.6637 Ops/s $\color{#35bf28}+1.51\%$
test_sac_speed[False-None] 10.4466ms 8.5129ms 117.4693 Ops/s 116.6924 Ops/s $\color{#35bf28}+0.67\%$
test_sac_speed[False-backward] 13.7023ms 11.7167ms 85.3486 Ops/s 85.4735 Ops/s $\color{#d91a1a}-0.15\%$
test_sac_speed[True-None] 2.6793ms 2.2486ms 444.7213 Ops/s 442.3315 Ops/s $\color{#35bf28}+0.54\%$
test_sac_speed[True-backward] 4.8047ms 4.0389ms 247.5952 Ops/s 216.5682 Ops/s $\textbf{\color{#35bf28}+14.33\%}$
test_sac_speed[reduce-overhead-None] 2.6882ms 2.1514ms 464.8170 Ops/s 386.3794 Ops/s $\textbf{\color{#35bf28}+20.30\%}$
test_sac_speed[reduce-overhead-backward] 5.3821ms 4.1489ms 241.0293 Ops/s 206.8565 Ops/s $\textbf{\color{#35bf28}+16.52\%}$
test_redq_speed[False-None] 15.9269ms 13.6681ms 73.1630 Ops/s 65.7553 Ops/s $\textbf{\color{#35bf28}+11.27\%}$
test_redq_speed[False-backward] 25.1705ms 23.4222ms 42.6945 Ops/s 39.9344 Ops/s $\textbf{\color{#35bf28}+6.91\%}$
test_redq_speed[True-None] 7.1319ms 6.5462ms 152.7596 Ops/s 154.7802 Ops/s $\color{#d91a1a}-1.31\%$
test_redq_speed[True-backward] 14.2133ms 13.8277ms 72.3185 Ops/s 68.6842 Ops/s $\textbf{\color{#35bf28}+5.29\%}$
test_redq_speed[reduce-overhead-None] 7.1937ms 6.2123ms 160.9699 Ops/s 149.5781 Ops/s $\textbf{\color{#35bf28}+7.62\%}$
test_redq_speed[reduce-overhead-backward] 16.3051ms 13.9601ms 71.6329 Ops/s 67.0696 Ops/s $\textbf{\color{#35bf28}+6.80\%}$
test_redq_deprec_speed[False-None] 16.7512ms 14.1916ms 70.4642 Ops/s 66.2359 Ops/s $\textbf{\color{#35bf28}+6.38\%}$
test_redq_deprec_speed[False-backward] 22.1543ms 20.3870ms 49.0508 Ops/s 46.8194 Ops/s $\color{#35bf28}+4.77\%$
test_redq_deprec_speed[True-None] 5.2488ms 4.5734ms 218.6573 Ops/s 203.9210 Ops/s $\textbf{\color{#35bf28}+7.23\%}$
test_redq_deprec_speed[True-backward] 11.4747ms 10.0466ms 99.5366 Ops/s 97.9819 Ops/s $\color{#35bf28}+1.59\%$
test_redq_deprec_speed[reduce-overhead-None] 5.2700ms 4.5870ms 218.0056 Ops/s 208.9419 Ops/s $\color{#35bf28}+4.34\%$
test_redq_deprec_speed[reduce-overhead-backward] 10.5016ms 9.9063ms 100.9456 Ops/s 98.4946 Ops/s $\color{#35bf28}+2.49\%$
test_td3_speed[False-None] 9.3327ms 8.8541ms 112.9417 Ops/s 113.5376 Ops/s $\color{#d91a1a}-0.52\%$
test_td3_speed[False-backward] 14.4135ms 11.8847ms 84.1416 Ops/s 85.7305 Ops/s $\color{#d91a1a}-1.85\%$
test_td3_speed[True-None] 2.2628ms 1.8911ms 528.8043 Ops/s 515.0162 Ops/s $\color{#35bf28}+2.68\%$
test_td3_speed[True-backward] 4.7063ms 3.9606ms 252.4847 Ops/s 236.4460 Ops/s $\textbf{\color{#35bf28}+6.78\%}$
test_td3_speed[reduce-overhead-None] 2.2808ms 1.9490ms 513.0839 Ops/s 448.7434 Ops/s $\textbf{\color{#35bf28}+14.34\%}$
test_td3_speed[reduce-overhead-backward] 4.2052ms 3.6477ms 274.1479 Ops/s 224.4347 Ops/s $\textbf{\color{#35bf28}+22.15\%}$
test_cql_speed[False-None] 42.7955ms 39.0221ms 25.6265 Ops/s 24.7643 Ops/s $\color{#35bf28}+3.48\%$
test_cql_speed[False-backward] 52.0522ms 49.1637ms 20.3402 Ops/s 19.9744 Ops/s $\color{#35bf28}+1.83\%$
test_cql_speed[True-None] 20.0385ms 16.9416ms 59.0263 Ops/s 58.0237 Ops/s $\color{#35bf28}+1.73\%$
test_cql_speed[True-backward] 25.9697ms 24.4774ms 40.8540 Ops/s 39.5390 Ops/s $\color{#35bf28}+3.33\%$
test_cql_speed[reduce-overhead-None] 18.8921ms 17.1794ms 58.2091 Ops/s 57.8676 Ops/s $\color{#35bf28}+0.59\%$
test_cql_speed[reduce-overhead-backward] 26.8767ms 24.6263ms 40.6071 Ops/s 40.4335 Ops/s $\color{#35bf28}+0.43\%$
test_a2c_speed[False-None] 8.5107ms 7.7499ms 129.0333 Ops/s 126.3306 Ops/s $\color{#35bf28}+2.14\%$
test_a2c_speed[False-backward] 17.2168ms 15.9939ms 62.5239 Ops/s 62.9216 Ops/s $\color{#d91a1a}-0.63\%$
test_a2c_speed[True-None] 4.7278ms 4.0928ms 244.3337 Ops/s 241.5938 Ops/s $\color{#35bf28}+1.13\%$
test_a2c_speed[True-backward] 11.8533ms 11.1962ms 89.3158 Ops/s 85.1431 Ops/s $\color{#35bf28}+4.90\%$
test_a2c_speed[reduce-overhead-None] 6.1293ms 3.9545ms 252.8754 Ops/s 225.8467 Ops/s $\textbf{\color{#35bf28}+11.97\%}$
test_a2c_speed[reduce-overhead-backward] 12.4329ms 11.3145ms 88.3822 Ops/s 83.8040 Ops/s $\textbf{\color{#35bf28}+5.46\%}$
test_ppo_speed[False-None] 9.6420ms 8.0488ms 124.2422 Ops/s 115.8039 Ops/s $\textbf{\color{#35bf28}+7.29\%}$
test_ppo_speed[False-backward] 18.3605ms 16.2966ms 61.3624 Ops/s 59.7239 Ops/s $\color{#35bf28}+2.74\%$
test_ppo_speed[True-None] 4.9522ms 4.4254ms 225.9708 Ops/s 220.3072 Ops/s $\color{#35bf28}+2.57\%$
test_ppo_speed[True-backward] 11.8662ms 11.1441ms 89.7335 Ops/s 88.5093 Ops/s $\color{#35bf28}+1.38\%$
test_ppo_speed[reduce-overhead-None] 6.2041ms 4.4802ms 223.2059 Ops/s 205.7597 Ops/s $\textbf{\color{#35bf28}+8.48\%}$
test_ppo_speed[reduce-overhead-backward] 11.6070ms 10.8909ms 91.8198 Ops/s 90.2747 Ops/s $\color{#35bf28}+1.71\%$
test_reinforce_speed[False-None] 8.3030ms 6.8676ms 145.6106 Ops/s 143.1763 Ops/s $\color{#35bf28}+1.70\%$
test_reinforce_speed[False-backward] 12.6380ms 10.4474ms 95.7180 Ops/s 94.4739 Ops/s $\color{#35bf28}+1.32\%$
test_reinforce_speed[True-None] 3.9531ms 3.2766ms 305.1980 Ops/s 283.6935 Ops/s $\textbf{\color{#35bf28}+7.58\%}$
test_reinforce_speed[True-backward] 11.0595ms 9.9375ms 100.6293 Ops/s 95.8309 Ops/s $\textbf{\color{#35bf28}+5.01\%}$
test_reinforce_speed[reduce-overhead-None] 4.0733ms 3.3294ms 300.3523 Ops/s 286.9206 Ops/s $\color{#35bf28}+4.68\%$
test_reinforce_speed[reduce-overhead-backward] 10.4665ms 9.9281ms 100.7240 Ops/s 98.7550 Ops/s $\color{#35bf28}+1.99\%$
test_iql_speed[False-None] 35.9105ms 33.4531ms 29.8926 Ops/s 29.3646 Ops/s $\color{#35bf28}+1.80\%$
test_iql_speed[False-backward] 49.5744ms 46.9059ms 21.3193 Ops/s 20.9615 Ops/s $\color{#35bf28}+1.71\%$
test_iql_speed[True-None] 13.7293ms 12.0794ms 82.7857 Ops/s 79.2216 Ops/s $\color{#35bf28}+4.50\%$
test_iql_speed[True-backward] 25.6593ms 23.9957ms 41.6741 Ops/s 41.1441 Ops/s $\color{#35bf28}+1.29\%$
test_iql_speed[reduce-overhead-None] 13.2535ms 12.1439ms 82.3458 Ops/s 80.6130 Ops/s $\color{#35bf28}+2.15\%$
test_iql_speed[reduce-overhead-backward] 25.5916ms 24.0292ms 41.6160 Ops/s 41.1332 Ops/s $\color{#35bf28}+1.17\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.2664ms 5.3859ms 185.6704 Ops/s 187.2406 Ops/s $\color{#d91a1a}-0.84\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.6275ms 0.5695ms 1.7558 KOps/s 1.7765 KOps/s $\color{#d91a1a}-1.16\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8892ms 0.5370ms 1.8620 KOps/s 1.8003 KOps/s $\color{#35bf28}+3.43\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.7549ms 5.0785ms 196.9071 Ops/s 200.0462 Ops/s $\color{#d91a1a}-1.57\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0556ms 0.5539ms 1.8054 KOps/s 1.8320 KOps/s $\color{#d91a1a}-1.45\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9206ms 0.5298ms 1.8876 KOps/s 1.8652 KOps/s $\color{#35bf28}+1.20\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.7022ms 1.7240ms 580.0308 Ops/s 574.6960 Ops/s $\color{#35bf28}+0.93\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.4935ms 1.6442ms 608.1906 Ops/s 600.3335 Ops/s $\color{#35bf28}+1.31\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.8303ms 5.1605ms 193.7815 Ops/s 191.1795 Ops/s $\color{#35bf28}+1.36\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1054ms 0.7019ms 1.4247 KOps/s 1.4365 KOps/s $\color{#d91a1a}-0.82\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0440ms 0.6807ms 1.4691 KOps/s 1.4965 KOps/s $\color{#d91a1a}-1.83\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.3796ms 5.0025ms 199.8992 Ops/s 187.9726 Ops/s $\textbf{\color{#35bf28}+6.34\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1587ms 0.5637ms 1.7739 KOps/s 1.7988 KOps/s $\color{#d91a1a}-1.39\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8035ms 0.5380ms 1.8588 KOps/s 1.8678 KOps/s $\color{#d91a1a}-0.48\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.6380ms 5.1435ms 194.4192 Ops/s 195.4520 Ops/s $\color{#d91a1a}-0.53\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.4069ms 0.5567ms 1.7964 KOps/s 1.8049 KOps/s $\color{#d91a1a}-0.47\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8111ms 0.5292ms 1.8895 KOps/s 1.8959 KOps/s $\color{#d91a1a}-0.34\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0325ms 5.3804ms 185.8581 Ops/s 185.2455 Ops/s $\color{#35bf28}+0.33\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1092ms 0.7012ms 1.4261 KOps/s 1.4153 KOps/s $\color{#35bf28}+0.76\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.1862ms 0.6759ms 1.4796 KOps/s 1.4941 KOps/s $\color{#d91a1a}-0.97\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.8886ms 4.6870ms 213.3543 Ops/s 197.1562 Ops/s $\textbf{\color{#35bf28}+8.22\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 5.5756ms 2.3484ms 425.8160 Ops/s 408.0266 Ops/s $\color{#35bf28}+4.36\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.7734ms 1.5044ms 664.7275 Ops/s 631.7180 Ops/s $\textbf{\color{#35bf28}+5.23\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.5574s 15.5773ms 64.1962 Ops/s 204.8372 Ops/s $\textbf{\color{#d91a1a}-68.66\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.6891ms 2.3532ms 424.9490 Ops/s 412.7845 Ops/s $\color{#35bf28}+2.95\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 9.0399ms 1.6152ms 619.1102 Ops/s 686.3551 Ops/s $\textbf{\color{#d91a1a}-9.80\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.8144ms 4.7404ms 210.9524 Ops/s 209.4248 Ops/s $\color{#35bf28}+0.73\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.8842ms 2.5249ms 396.0616 Ops/s 389.0223 Ops/s $\color{#35bf28}+1.81\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 5.6383ms 1.5519ms 644.3859 Ops/s 564.0658 Ops/s $\textbf{\color{#35bf28}+14.24\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.0913ms 12.6357ms 79.1409 Ops/s 77.4449 Ops/s $\color{#35bf28}+2.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 23.2034ms 14.8797ms 67.2055 Ops/s 67.5793 Ops/s $\color{#d91a1a}-0.55\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 23.5215ms 21.4861ms 46.5417 Ops/s 46.2994 Ops/s $\color{#35bf28}+0.52\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.5271ms 14.9575ms 66.8560 Ops/s 66.5782 Ops/s $\color{#35bf28}+0.42\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 24.0220ms 21.4052ms 46.7177 Ops/s 45.9848 Ops/s $\color{#35bf28}+1.59\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 22.4228ms 16.3922ms 61.0047 Ops/s 60.8537 Ops/s $\color{#35bf28}+0.25\%$

[ghstack-poisoned]
@vmoens vmoens merged commit 1541dd5 into gh/vmoens/98/base Mar 3, 2025
68 of 72 checks passed
vmoens added a commit that referenced this pull request Mar 3, 2025
ghstack-source-id: cf1942ece8dfbd6506f91939561df7443bd840ab
Pull Request resolved: #2821
@vmoens vmoens deleted the gh/vmoens/98/head branch March 3, 2025 15:50
@vmoens vmoens added the enhancement New feature or request label Mar 3, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants