Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 22, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Oct 22, 2025
ghstack-source-id: 5704ab4
Pull-Request: #3218
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 22, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3218

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

❌ 11 New Failures, 1 Cancelled Job, 2 Unrelated Failures

As of commit f73be7c with merge base 47ad9d8 (image):

NEW FAILURES - The following jobs have failed:

CANCELLED JOB - The following job was cancelled. Please retry:

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

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

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

@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 Oct 22, 2025
@vmoens vmoens added the bug Something isn't working label Oct 22, 2025
@github-actions
Copy link

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 85.0032μs 82.4850μs 12.1234 KOps/s 11.8018 KOps/s $\color{#35bf28}+2.73\%$
test_tensor_to_bytestream_speed[torch.save] 0.1421ms 0.1408ms 7.1034 KOps/s 7.2671 KOps/s $\color{#d91a1a}-2.25\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1137s 0.1135s 8.8093 Ops/s 8.9105 Ops/s $\color{#d91a1a}-1.14\%$
test_tensor_to_bytestream_speed[numpy] 2.8791μs 2.8685μs 348.6132 KOps/s 346.7740 KOps/s $\color{#35bf28}+0.53\%$
test_tensor_to_bytestream_speed[safetensors] 42.1956μs 40.9853μs 24.3990 KOps/s 24.4539 KOps/s $\color{#d91a1a}-0.22\%$
test_simple 0.5334s 0.5321s 1.8792 Ops/s 1.7880 Ops/s $\textbf{\color{#35bf28}+5.11\%}$
test_transformed 1.1963s 1.1061s 0.9041 Ops/s 0.9029 Ops/s $\color{#35bf28}+0.13\%$
test_serial 1.6243s 1.6143s 0.6195 Ops/s 0.6098 Ops/s $\color{#35bf28}+1.59\%$
test_parallel 1.1920s 1.1323s 0.8832 Ops/s 0.9824 Ops/s $\textbf{\color{#d91a1a}-10.10\%}$
test_step_mdp_speed[True-True-True-True-True] 89.5920μs 42.7857μs 23.3723 KOps/s 22.5598 KOps/s $\color{#35bf28}+3.60\%$
test_step_mdp_speed[True-True-True-True-False] 56.9010μs 24.6265μs 40.6067 KOps/s 40.7505 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[True-True-True-False-True] 73.0710μs 24.7840μs 40.3485 KOps/s 40.7250 KOps/s $\color{#d91a1a}-0.92\%$
test_step_mdp_speed[True-True-True-False-False] 54.9010μs 13.4968μs 74.0917 KOps/s 73.7148 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-True-False-True-True] 82.3020μs 46.9125μs 21.3163 KOps/s 21.2343 KOps/s $\color{#35bf28}+0.39\%$
test_step_mdp_speed[True-True-False-True-False] 59.7720μs 26.9894μs 37.0516 KOps/s 36.7041 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[True-True-False-False-True] 61.3620μs 27.2467μs 36.7018 KOps/s 36.5313 KOps/s $\color{#35bf28}+0.47\%$
test_step_mdp_speed[True-True-False-False-False] 52.5210μs 16.2862μs 61.4017 KOps/s 61.7266 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[True-False-True-True-True] 93.9920μs 49.4958μs 20.2037 KOps/s 20.2870 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[True-False-True-True-False] 67.8120μs 30.5439μs 32.7398 KOps/s 33.5464 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[True-False-True-False-True] 79.4310μs 27.5280μs 36.3267 KOps/s 36.2165 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[True-False-True-False-False] 55.1410μs 16.1341μs 61.9806 KOps/s 61.9771 KOps/s $+0.01\%$
test_step_mdp_speed[True-False-False-True-True] 92.9320μs 52.1430μs 19.1780 KOps/s 19.0476 KOps/s $\color{#35bf28}+0.68\%$
test_step_mdp_speed[True-False-False-True-False] 66.7710μs 32.4541μs 30.8127 KOps/s 30.5068 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[True-False-False-False-True] 63.0710μs 29.8039μs 33.5526 KOps/s 33.9341 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[True-False-False-False-False] 51.1510μs 18.7947μs 53.2065 KOps/s 53.5124 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[False-True-True-True-True] 95.0520μs 49.2822μs 20.2913 KOps/s 20.1100 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[False-True-True-True-False] 72.8110μs 30.1563μs 33.1606 KOps/s 33.3987 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[False-True-True-False-True] 2.4407ms 31.7300μs 31.5159 KOps/s 32.3258 KOps/s $\color{#d91a1a}-2.51\%$
test_step_mdp_speed[False-True-True-False-False] 80.8820μs 17.8384μs 56.0587 KOps/s 57.0527 KOps/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[False-True-False-True-True] 91.9220μs 51.4031μs 19.4541 KOps/s 19.4748 KOps/s $\color{#d91a1a}-0.11\%$
test_step_mdp_speed[False-True-False-True-False] 64.4810μs 32.2729μs 30.9857 KOps/s 30.8500 KOps/s $\color{#35bf28}+0.44\%$
test_step_mdp_speed[False-True-False-False-True] 67.8510μs 33.6619μs 29.7072 KOps/s 29.8661 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[False-True-False-False-False] 44.3910μs 20.3910μs 49.0412 KOps/s 49.1077 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[False-False-True-True-True] 97.9320μs 55.7349μs 17.9421 KOps/s 18.6888 KOps/s $\color{#d91a1a}-4.00\%$
test_step_mdp_speed[False-False-True-True-False] 62.4420μs 35.4207μs 28.2320 KOps/s 28.8776 KOps/s $\color{#d91a1a}-2.24\%$
test_step_mdp_speed[False-False-True-False-True] 69.9820μs 34.4665μs 29.0136 KOps/s 30.0397 KOps/s $\color{#d91a1a}-3.42\%$
test_step_mdp_speed[False-False-True-False-False] 55.2810μs 20.3647μs 49.1045 KOps/s 49.7393 KOps/s $\color{#d91a1a}-1.28\%$
test_step_mdp_speed[False-False-False-True-True] 89.6120μs 57.1337μs 17.5028 KOps/s 17.9474 KOps/s $\color{#d91a1a}-2.48\%$
test_step_mdp_speed[False-False-False-True-False] 76.0420μs 37.7207μs 26.5106 KOps/s 26.8437 KOps/s $\color{#d91a1a}-1.24\%$
test_step_mdp_speed[False-False-False-False-True] 0.1085ms 36.6010μs 27.3217 KOps/s 27.9341 KOps/s $\color{#d91a1a}-2.19\%$
test_step_mdp_speed[False-False-False-False-False] 53.2020μs 22.9578μs 43.5581 KOps/s 43.2984 KOps/s $\color{#35bf28}+0.60\%$
test_values[generalized_advantage_estimate-True-True] 9.7870ms 9.4955ms 105.3127 Ops/s 106.2962 Ops/s $\color{#d91a1a}-0.93\%$
test_values[vec_generalized_advantage_estimate-True-True] 21.6424ms 17.7235ms 56.4221 Ops/s 90.3174 Ops/s $\textbf{\color{#d91a1a}-37.53\%}$
test_values[td0_return_estimate-False-False] 0.2980ms 0.1349ms 7.4129 KOps/s 7.8762 KOps/s $\textbf{\color{#d91a1a}-5.88\%}$
test_values[td1_return_estimate-False-False] 27.7583ms 26.1602ms 38.2261 Ops/s 38.5628 Ops/s $\color{#d91a1a}-0.87\%$
test_values[vec_td1_return_estimate-False-False] 21.4266ms 17.8748ms 55.9448 Ops/s 89.8916 Ops/s $\textbf{\color{#d91a1a}-37.76\%}$
test_values[td_lambda_return_estimate-True-False] 40.3334ms 38.6501ms 25.8732 Ops/s 26.1354 Ops/s $\color{#d91a1a}-1.00\%$
test_values[vec_td_lambda_return_estimate-True-False] 21.8060ms 17.8430ms 56.0445 Ops/s 89.9909 Ops/s $\textbf{\color{#d91a1a}-37.72\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.3048ms 8.1520ms 122.6691 Ops/s 124.0245 Ops/s $\color{#d91a1a}-1.09\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8865ms 1.5304ms 653.4250 Ops/s 671.9587 Ops/s $\color{#d91a1a}-2.76\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4473ms 0.4031ms 2.4808 KOps/s 2.5501 KOps/s $\color{#d91a1a}-2.72\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 38.9723ms 35.3250ms 28.3085 Ops/s 40.9310 Ops/s $\textbf{\color{#d91a1a}-30.84\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.1353ms 1.6873ms 592.6655 Ops/s 590.2427 Ops/s $\color{#35bf28}+0.41\%$
test_dqn_speed[False-None] 1.5091ms 1.3906ms 719.1039 Ops/s 725.7117 Ops/s $\color{#d91a1a}-0.91\%$
test_dqn_speed[False-backward] 1.9565ms 1.8844ms 530.6617 Ops/s 541.2884 Ops/s $\color{#d91a1a}-1.96\%$
test_dqn_speed[True-None] 0.6381ms 0.5144ms 1.9442 KOps/s 1.9345 KOps/s $\color{#35bf28}+0.50\%$
test_dqn_speed[True-backward] 0.9800ms 0.9410ms 1.0627 KOps/s 927.9203 Ops/s $\textbf{\color{#35bf28}+14.52\%}$
test_dqn_speed[reduce-overhead-None] 0.6039ms 0.5069ms 1.9729 KOps/s 2.0310 KOps/s $\color{#d91a1a}-2.86\%$
test_dqn_speed[reduce-overhead-backward] 0.9790ms 0.9319ms 1.0731 KOps/s 927.3266 Ops/s $\textbf{\color{#35bf28}+15.72\%}$
test_ddpg_speed[False-None] 3.2362ms 2.8593ms 349.7373 Ops/s 355.3767 Ops/s $\color{#d91a1a}-1.59\%$
test_ddpg_speed[False-backward] 4.2942ms 4.0068ms 249.5773 Ops/s 252.2403 Ops/s $\color{#d91a1a}-1.06\%$
test_ddpg_speed[True-None] 1.4709ms 1.3532ms 739.0000 Ops/s 720.5274 Ops/s $\color{#35bf28}+2.56\%$
test_ddpg_speed[True-backward] 2.3531ms 2.3005ms 434.6906 Ops/s 379.2588 Ops/s $\textbf{\color{#35bf28}+14.62\%}$
test_ddpg_speed[reduce-overhead-None] 1.4759ms 1.3455ms 743.2332 Ops/s 724.1470 Ops/s $\color{#35bf28}+2.64\%$
test_ddpg_speed[reduce-overhead-backward] 2.3255ms 2.2901ms 436.6577 Ops/s 430.4832 Ops/s $\color{#35bf28}+1.43\%$
test_sac_speed[False-None] 8.1117ms 7.6082ms 131.4365 Ops/s 130.9126 Ops/s $\color{#35bf28}+0.40\%$
test_sac_speed[False-backward] 11.1842ms 10.7769ms 92.7911 Ops/s 92.4846 Ops/s $\color{#35bf28}+0.33\%$
test_sac_speed[True-None] 2.2429ms 2.0745ms 482.0343 Ops/s 463.2960 Ops/s $\color{#35bf28}+4.04\%$
test_sac_speed[True-backward] 4.0125ms 3.9223ms 254.9516 Ops/s 228.2623 Ops/s $\textbf{\color{#35bf28}+11.69\%}$
test_sac_speed[reduce-overhead-None] 2.2999ms 2.0777ms 481.3012 Ops/s 461.8013 Ops/s $\color{#35bf28}+4.22\%$
test_sac_speed[reduce-overhead-backward] 4.0375ms 3.9331ms 254.2530 Ops/s 218.5258 Ops/s $\textbf{\color{#35bf28}+16.35\%}$
test_redq_speed[False-None] 15.3684ms 10.3716ms 96.4169 Ops/s 99.0606 Ops/s $\color{#d91a1a}-2.67\%$
test_redq_speed[False-backward] 22.9363ms 17.9366ms 55.7520 Ops/s 57.8647 Ops/s $\color{#d91a1a}-3.65\%$
test_redq_speed[True-None] 4.5214ms 4.2610ms 234.6890 Ops/s 229.0759 Ops/s $\color{#35bf28}+2.45\%$
test_redq_speed[True-backward] 10.0502ms 9.7888ms 102.1572 Ops/s 100.8269 Ops/s $\color{#35bf28}+1.32\%$
test_redq_speed[reduce-overhead-None] 4.5160ms 4.2987ms 232.6305 Ops/s 226.4301 Ops/s $\color{#35bf28}+2.74\%$
test_redq_speed[reduce-overhead-backward] 10.1034ms 9.8998ms 101.0126 Ops/s 102.8605 Ops/s $\color{#d91a1a}-1.80\%$
test_redq_deprec_speed[False-None] 11.5170ms 10.7350ms 93.1533 Ops/s 93.7560 Ops/s $\color{#d91a1a}-0.64\%$
test_redq_deprec_speed[False-backward] 16.3257ms 15.6455ms 63.9161 Ops/s 65.7707 Ops/s $\color{#d91a1a}-2.82\%$
test_redq_deprec_speed[True-None] 3.7708ms 3.5412ms 282.3934 Ops/s 282.4856 Ops/s $\color{#d91a1a}-0.03\%$
test_redq_deprec_speed[True-backward] 7.6995ms 7.4817ms 133.6593 Ops/s 132.6976 Ops/s $\color{#35bf28}+0.72\%$
test_redq_deprec_speed[reduce-overhead-None] 3.6853ms 3.5010ms 285.6305 Ops/s 278.8120 Ops/s $\color{#35bf28}+2.45\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.6075ms 7.4267ms 134.6493 Ops/s 133.9686 Ops/s $\color{#35bf28}+0.51\%$
test_td3_speed[False-None] 7.9442ms 7.6988ms 129.8909 Ops/s 130.2021 Ops/s $\color{#d91a1a}-0.24\%$
test_td3_speed[False-backward] 11.0615ms 10.5018ms 95.2221 Ops/s 95.8518 Ops/s $\color{#d91a1a}-0.66\%$
test_td3_speed[True-None] 1.7815ms 1.7438ms 573.4588 Ops/s 565.6243 Ops/s $\color{#35bf28}+1.39\%$
test_td3_speed[True-backward] 3.7878ms 3.5400ms 282.4825 Ops/s 241.1019 Ops/s $\textbf{\color{#35bf28}+17.16\%}$
test_td3_speed[reduce-overhead-None] 1.7718ms 1.7234ms 580.2415 Ops/s 570.2488 Ops/s $\color{#35bf28}+1.75\%$
test_td3_speed[reduce-overhead-backward] 3.6015ms 3.5225ms 283.8881 Ops/s 245.7627 Ops/s $\textbf{\color{#35bf28}+15.51\%}$
test_cql_speed[False-None] 29.9516ms 25.5754ms 39.1001 Ops/s 39.5105 Ops/s $\color{#d91a1a}-1.04\%$
test_cql_speed[False-backward] 37.7377ms 34.6267ms 28.8794 Ops/s 29.1747 Ops/s $\color{#d91a1a}-1.01\%$
test_cql_speed[True-None] 12.6327ms 12.2427ms 81.6816 Ops/s 80.5269 Ops/s $\color{#35bf28}+1.43\%$
test_cql_speed[True-backward] 18.7433ms 18.1913ms 54.9712 Ops/s 56.1014 Ops/s $\color{#d91a1a}-2.01\%$
test_cql_speed[reduce-overhead-None] 12.4676ms 12.1488ms 82.3124 Ops/s 81.7003 Ops/s $\color{#35bf28}+0.75\%$
test_cql_speed[reduce-overhead-backward] 18.5378ms 18.1225ms 55.1799 Ops/s 55.9366 Ops/s $\color{#d91a1a}-1.35\%$
test_a2c_speed[False-None] 5.7194ms 5.3812ms 185.8309 Ops/s 188.7236 Ops/s $\color{#d91a1a}-1.53\%$
test_a2c_speed[False-backward] 11.9813ms 11.7613ms 85.0246 Ops/s 84.9608 Ops/s $\color{#35bf28}+0.08\%$
test_a2c_speed[True-None] 3.7699ms 3.6451ms 274.3379 Ops/s 272.2731 Ops/s $\color{#35bf28}+0.76\%$
test_a2c_speed[True-backward] 8.7815ms 8.6234ms 115.9635 Ops/s 118.1160 Ops/s $\color{#d91a1a}-1.82\%$
test_a2c_speed[reduce-overhead-None] 3.8136ms 3.6631ms 272.9937 Ops/s 276.1064 Ops/s $\color{#d91a1a}-1.13\%$
test_a2c_speed[reduce-overhead-backward] 8.8949ms 8.6961ms 114.9940 Ops/s 114.2766 Ops/s $\color{#35bf28}+0.63\%$
test_ppo_speed[False-None] 5.9682ms 5.7864ms 172.8185 Ops/s 174.9198 Ops/s $\color{#d91a1a}-1.20\%$
test_ppo_speed[False-backward] 12.7019ms 12.3287ms 81.1117 Ops/s 81.7650 Ops/s $\color{#d91a1a}-0.80\%$
test_ppo_speed[True-None] 3.7494ms 3.6119ms 276.8658 Ops/s 277.8817 Ops/s $\color{#d91a1a}-0.37\%$
test_ppo_speed[True-backward] 8.7213ms 8.4190ms 118.7786 Ops/s 119.0503 Ops/s $\color{#d91a1a}-0.23\%$
test_ppo_speed[reduce-overhead-None] 3.7092ms 3.5803ms 279.3093 Ops/s 278.4608 Ops/s $\color{#35bf28}+0.30\%$
test_ppo_speed[reduce-overhead-backward] 8.8572ms 8.6317ms 115.8522 Ops/s 115.3657 Ops/s $\color{#35bf28}+0.42\%$
test_reinforce_speed[False-None] 4.7315ms 4.4731ms 223.5586 Ops/s 223.2528 Ops/s $\color{#35bf28}+0.14\%$
test_reinforce_speed[False-backward] 7.5508ms 7.2942ms 137.0957 Ops/s 138.4414 Ops/s $\color{#d91a1a}-0.97\%$
test_reinforce_speed[True-None] 3.0531ms 2.8107ms 355.7833 Ops/s 325.3679 Ops/s $\textbf{\color{#35bf28}+9.35\%}$
test_reinforce_speed[True-backward] 7.8360ms 7.6419ms 130.8580 Ops/s 131.2532 Ops/s $\color{#d91a1a}-0.30\%$
test_reinforce_speed[reduce-overhead-None] 3.0312ms 2.7949ms 357.7940 Ops/s 349.3340 Ops/s $\color{#35bf28}+2.42\%$
test_reinforce_speed[reduce-overhead-backward] 8.0514ms 7.8254ms 127.7894 Ops/s 121.3978 Ops/s $\textbf{\color{#35bf28}+5.26\%}$
test_iql_speed[False-None] 20.0974ms 19.3882ms 51.5777 Ops/s 51.8051 Ops/s $\color{#d91a1a}-0.44\%$
test_iql_speed[False-backward] 35.6414ms 29.9977ms 33.3359 Ops/s 33.4634 Ops/s $\color{#d91a1a}-0.38\%$
test_iql_speed[True-None] 8.7784ms 8.3962ms 119.1008 Ops/s 117.7479 Ops/s $\color{#35bf28}+1.15\%$
test_iql_speed[True-backward] 17.2009ms 16.6285ms 60.1376 Ops/s 60.2946 Ops/s $\color{#d91a1a}-0.26\%$
test_iql_speed[reduce-overhead-None] 10.9623ms 8.7845ms 113.8366 Ops/s 119.2838 Ops/s $\color{#d91a1a}-4.57\%$
test_iql_speed[reduce-overhead-backward] 17.4126ms 17.0793ms 58.5504 Ops/s 59.4811 Ops/s $\color{#d91a1a}-1.56\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.3795ms 5.8595ms 170.6637 Ops/s 171.5341 Ops/s $\color{#d91a1a}-0.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6696ms 0.3495ms 2.8609 KOps/s 3.3519 KOps/s $\textbf{\color{#d91a1a}-14.65\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6353ms 0.3710ms 2.6957 KOps/s 3.4226 KOps/s $\textbf{\color{#d91a1a}-21.24\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.8608ms 5.6530ms 176.8976 Ops/s 179.4029 Ops/s $\color{#d91a1a}-1.40\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5754s 0.7600ms 1.3158 KOps/s 2.8941 KOps/s $\textbf{\color{#d91a1a}-54.54\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5965ms 0.3067ms 3.2605 KOps/s 3.0638 KOps/s $\textbf{\color{#35bf28}+6.42\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5752ms 1.3140ms 761.0491 Ops/s 746.3646 Ops/s $\color{#35bf28}+1.97\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4757ms 1.2193ms 820.1556 Ops/s 795.0999 Ops/s $\color{#35bf28}+3.15\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.8977ms 5.7921ms 172.6476 Ops/s 175.2102 Ops/s $\color{#d91a1a}-1.46\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9315ms 0.4377ms 2.2846 KOps/s 2.0953 KOps/s $\textbf{\color{#35bf28}+9.04\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6262ms 0.4220ms 2.3694 KOps/s 2.1605 KOps/s $\textbf{\color{#35bf28}+9.67\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.7908ms 5.6475ms 177.0705 Ops/s 178.3395 Ops/s $\color{#d91a1a}-0.71\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8564ms 0.3709ms 2.6962 KOps/s 2.8509 KOps/s $\textbf{\color{#d91a1a}-5.43\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5358ms 0.3440ms 2.9070 KOps/s 3.0133 KOps/s $\color{#d91a1a}-3.53\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9245ms 5.5999ms 178.5743 Ops/s 178.4353 Ops/s $\color{#35bf28}+0.08\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0353ms 0.2837ms 3.5250 KOps/s 2.8931 KOps/s $\textbf{\color{#35bf28}+21.84\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4651ms 0.2857ms 3.5001 KOps/s 3.0729 KOps/s $\textbf{\color{#35bf28}+13.90\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 9.5637ms 5.9194ms 168.9353 Ops/s 171.9653 Ops/s $\color{#d91a1a}-1.76\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.2722ms 0.4866ms 2.0552 KOps/s 2.0597 KOps/s $\color{#d91a1a}-0.22\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7118ms 0.4613ms 2.1679 KOps/s 2.1521 KOps/s $\color{#35bf28}+0.74\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.4955s 14.7982ms 67.5759 Ops/s 197.1395 Ops/s $\textbf{\color{#d91a1a}-65.72\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.1671ms 2.0247ms 493.8987 Ops/s 439.2724 Ops/s $\textbf{\color{#35bf28}+12.44\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4268ms 1.1510ms 868.8153 Ops/s 884.0174 Ops/s $\color{#d91a1a}-1.72\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.6021ms 4.9745ms 201.0241 Ops/s 56.4822 Ops/s $\textbf{\color{#35bf28}+255.91\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 10.8128ms 2.0448ms 489.0414 Ops/s 603.8140 Ops/s $\textbf{\color{#d91a1a}-19.01\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 3.3409ms 1.0826ms 923.7063 Ops/s 802.0897 Ops/s $\textbf{\color{#35bf28}+15.16\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.4638s 14.3514ms 69.6794 Ops/s 189.2501 Ops/s $\textbf{\color{#d91a1a}-63.18\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.1828ms 2.0972ms 476.8199 Ops/s 468.1205 Ops/s $\color{#35bf28}+1.86\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.5827ms 1.2880ms 776.3923 Ops/s 738.3674 Ops/s $\textbf{\color{#35bf28}+5.15\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.4030ms 32.6659ms 30.6130 Ops/s 30.7841 Ops/s $\color{#d91a1a}-0.56\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 18.2729ms 16.9201ms 59.1014 Ops/s 59.5363 Ops/s $\color{#d91a1a}-0.73\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 34.8961ms 33.0935ms 30.2174 Ops/s 29.9321 Ops/s $\color{#35bf28}+0.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.8904ms 17.0907ms 58.5114 Ops/s 58.6090 Ops/s $\color{#d91a1a}-0.17\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 36.2069ms 34.5194ms 28.9692 Ops/s 28.3547 Ops/s $\color{#35bf28}+2.17\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.3872ms 18.4155ms 54.3022 Ops/s 53.8128 Ops/s $\color{#35bf28}+0.91\%$

vmoens added a commit that referenced this pull request Oct 23, 2025
ghstack-source-id: 5704ab4
Pull-Request: #3218
@vmoens vmoens merged commit f73be7c into gh/vmoens/168/base Oct 23, 2025
87 of 101 checks passed
@vmoens vmoens deleted the gh/vmoens/168/head branch October 23, 2025 00:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

bug Something isn't working 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