Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Oct 14, 2025

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Oct 14, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3191

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

❌ 12 New Failures, 1 Unrelated Failure

As of commit 7a28716 with merge base 3d1748f (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following job 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.

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens mentioned this pull request Oct 18, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens added the enhancement New feature or request label Oct 20, 2025
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@github-actions
Copy link

github-actions bot commented Oct 20, 2025

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

Total Benchmarks: 154. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.9285μs 82.2983μs 12.1509 KOps/s 12.0082 KOps/s $\color{#35bf28}+1.19\%$
test_tensor_to_bytestream_speed[torch.save] 0.1477ms 0.1448ms 6.9066 KOps/s 6.9305 KOps/s $\color{#d91a1a}-0.34\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1174s 0.1172s 8.5327 Ops/s 7.5616 Ops/s $\textbf{\color{#35bf28}+12.84\%}$
test_tensor_to_bytestream_speed[numpy] 2.8369μs 2.8315μs 353.1709 KOps/s 343.4892 KOps/s $\color{#35bf28}+2.82\%$
test_tensor_to_bytestream_speed[safetensors] 44.5226μs 42.4520μs 23.5560 KOps/s 23.1044 KOps/s $\color{#35bf28}+1.95\%$
test_simple 0.5562s 0.5556s 1.8000 Ops/s 1.7094 Ops/s $\textbf{\color{#35bf28}+5.30\%}$
test_transformed 1.1256s 1.1234s 0.8901 Ops/s 0.8629 Ops/s $\color{#35bf28}+3.16\%$
test_serial 1.6939s 1.6862s 0.5931 Ops/s 0.5779 Ops/s $\color{#35bf28}+2.63\%$
test_parallel 1.1754s 1.1270s 0.8873 Ops/s 0.9191 Ops/s $\color{#d91a1a}-3.46\%$
test_step_mdp_speed[True-True-True-True-True] 0.2475ms 46.4553μs 21.5261 KOps/s 22.1035 KOps/s $\color{#d91a1a}-2.61\%$
test_step_mdp_speed[True-True-True-True-False] 57.1210μs 25.7257μs 38.8716 KOps/s 39.0670 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[True-True-True-False-True] 58.4110μs 26.0408μs 38.4012 KOps/s 39.0837 KOps/s $\color{#d91a1a}-1.75\%$
test_step_mdp_speed[True-True-True-False-False] 40.1310μs 14.3375μs 69.7473 KOps/s 70.5459 KOps/s $\color{#d91a1a}-1.13\%$
test_step_mdp_speed[True-True-False-True-True] 73.8910μs 49.2696μs 20.2965 KOps/s 20.5178 KOps/s $\color{#d91a1a}-1.08\%$
test_step_mdp_speed[True-True-False-True-False] 63.6610μs 28.6579μs 34.8944 KOps/s 34.6994 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-True-False-False-True] 91.9120μs 28.7679μs 34.7610 KOps/s 34.9129 KOps/s $\color{#d91a1a}-0.44\%$
test_step_mdp_speed[True-True-False-False-False] 44.4010μs 17.0631μs 58.6059 KOps/s 58.3972 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[True-False-True-True-True] 88.7220μs 52.1007μs 19.1936 KOps/s 19.1843 KOps/s $\color{#35bf28}+0.05\%$
test_step_mdp_speed[True-False-True-True-False] 63.5310μs 31.3934μs 31.8538 KOps/s 31.6027 KOps/s $\color{#35bf28}+0.79\%$
test_step_mdp_speed[True-False-True-False-True] 58.2010μs 29.0325μs 34.4441 KOps/s 35.2029 KOps/s $\color{#d91a1a}-2.16\%$
test_step_mdp_speed[True-False-True-False-False] 54.5810μs 17.0676μs 58.5906 KOps/s 59.1384 KOps/s $\color{#d91a1a}-0.93\%$
test_step_mdp_speed[True-False-False-True-True] 83.3210μs 54.6601μs 18.2949 KOps/s 18.4702 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[True-False-False-True-False] 72.1510μs 34.0530μs 29.3660 KOps/s 29.0566 KOps/s $\color{#35bf28}+1.06\%$
test_step_mdp_speed[True-False-False-False-True] 59.2110μs 31.9461μs 31.3027 KOps/s 31.9108 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[True-False-False-False-False] 50.7210μs 19.8773μs 50.3086 KOps/s 50.1146 KOps/s $\color{#35bf28}+0.39\%$
test_step_mdp_speed[False-True-True-True-True] 86.0810μs 52.2261μs 19.1475 KOps/s 19.3421 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-True-True-True-False] 55.3210μs 31.5932μs 31.6523 KOps/s 31.3499 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[False-True-True-False-True] 2.4647ms 33.3771μs 29.9606 KOps/s 31.0856 KOps/s $\color{#d91a1a}-3.62\%$
test_step_mdp_speed[False-True-True-False-False] 48.2810μs 19.2427μs 51.9678 KOps/s 52.0764 KOps/s $\color{#d91a1a}-0.21\%$
test_step_mdp_speed[False-True-False-True-True] 87.1220μs 55.1834μs 18.1214 KOps/s 18.4426 KOps/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[False-True-False-True-False] 71.8410μs 34.5278μs 28.9622 KOps/s 29.1211 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-True-False-False-True] 72.2910μs 35.6665μs 28.0375 KOps/s 28.4423 KOps/s $\color{#d91a1a}-1.42\%$
test_step_mdp_speed[False-True-False-False-False] 54.5910μs 21.6957μs 46.0921 KOps/s 45.8285 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[False-False-True-True-True] 98.4520μs 56.5106μs 17.6958 KOps/s 17.5727 KOps/s $\color{#35bf28}+0.70\%$
test_step_mdp_speed[False-False-True-True-False] 75.5510μs 37.2061μs 26.8773 KOps/s 26.9331 KOps/s $\color{#d91a1a}-0.21\%$
test_step_mdp_speed[False-False-True-False-True] 76.6210μs 35.3239μs 28.3094 KOps/s 28.5469 KOps/s $\color{#d91a1a}-0.83\%$
test_step_mdp_speed[False-False-True-False-False] 53.6910μs 21.6988μs 46.0855 KOps/s 45.8409 KOps/s $\color{#35bf28}+0.53\%$
test_step_mdp_speed[False-False-False-True-True] 90.0310μs 60.0179μs 16.6617 KOps/s 16.7871 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-False-False-True-False] 0.1083ms 39.1165μs 25.5647 KOps/s 24.9741 KOps/s $\color{#35bf28}+2.36\%$
test_step_mdp_speed[False-False-False-False-True] 65.8110μs 37.9014μs 26.3843 KOps/s 26.7077 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[False-False-False-False-False] 54.5610μs 24.1521μs 41.4042 KOps/s 40.8322 KOps/s $\color{#35bf28}+1.40\%$
test_values[generalized_advantage_estimate-True-True] 10.4827ms 10.2648ms 97.4203 Ops/s 97.5288 Ops/s $\color{#d91a1a}-0.11\%$
test_values[vec_generalized_advantage_estimate-True-True] 13.5373ms 11.1651ms 89.5647 Ops/s 89.6520 Ops/s $\color{#d91a1a}-0.10\%$
test_values[td0_return_estimate-False-False] 0.2531ms 0.1327ms 7.5330 KOps/s 7.4745 KOps/s $\color{#35bf28}+0.78\%$
test_values[td1_return_estimate-False-False] 28.6370ms 27.7397ms 36.0494 Ops/s 35.2580 Ops/s $\color{#35bf28}+2.24\%$
test_values[vec_td1_return_estimate-False-False] 11.9543ms 11.2620ms 88.7943 Ops/s 88.4732 Ops/s $\color{#35bf28}+0.36\%$
test_values[td_lambda_return_estimate-True-False] 43.5645ms 41.7931ms 23.9274 Ops/s 23.7736 Ops/s $\color{#35bf28}+0.65\%$
test_values[vec_td_lambda_return_estimate-True-False] 11.8347ms 11.2408ms 88.9616 Ops/s 88.6103 Ops/s $\color{#35bf28}+0.40\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.8169ms 8.7253ms 114.6089 Ops/s 111.4573 Ops/s $\color{#35bf28}+2.83\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.9528ms 1.5378ms 650.2951 Ops/s 671.0648 Ops/s $\color{#d91a1a}-3.10\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4822ms 0.4178ms 2.3932 KOps/s 2.3391 KOps/s $\color{#35bf28}+2.32\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 23.7641ms 23.1866ms 43.1284 Ops/s 37.9874 Ops/s $\textbf{\color{#35bf28}+13.53\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.2551ms 1.7682ms 565.5471 Ops/s 578.7037 Ops/s $\color{#d91a1a}-2.27\%$
test_dqn_speed[False-None] 6.4730ms 1.4408ms 694.0696 Ops/s 687.7528 Ops/s $\color{#35bf28}+0.92\%$
test_dqn_speed[False-backward] 2.0090ms 1.9410ms 515.1981 Ops/s 516.4193 Ops/s $\color{#d91a1a}-0.24\%$
test_dqn_speed[True-None] 0.9062ms 0.5114ms 1.9556 KOps/s 1.9067 KOps/s $\color{#35bf28}+2.56\%$
test_dqn_speed[True-backward] 1.0048ms 0.9688ms 1.0323 KOps/s 850.7924 Ops/s $\textbf{\color{#35bf28}+21.33\%}$
test_dqn_speed[reduce-overhead-None] 0.8812ms 0.5057ms 1.9774 KOps/s 1.8824 KOps/s $\textbf{\color{#35bf28}+5.05\%}$
test_dqn_speed[reduce-overhead-backward] 1.0077ms 0.9506ms 1.0520 KOps/s 1.0312 KOps/s $\color{#35bf28}+2.02\%$
test_ddpg_speed[False-None] 3.1966ms 2.9025ms 344.5279 Ops/s 342.3970 Ops/s $\color{#35bf28}+0.62\%$
test_ddpg_speed[False-backward] 4.2167ms 4.1212ms 242.6492 Ops/s 242.0704 Ops/s $\color{#35bf28}+0.24\%$
test_ddpg_speed[True-None] 1.7434ms 1.3794ms 724.9771 Ops/s 714.7922 Ops/s $\color{#35bf28}+1.42\%$
test_ddpg_speed[True-backward] 2.4118ms 2.3574ms 424.1958 Ops/s 403.8023 Ops/s $\textbf{\color{#35bf28}+5.05\%}$
test_ddpg_speed[reduce-overhead-None] 1.7383ms 1.3645ms 732.8441 Ops/s 719.9638 Ops/s $\color{#35bf28}+1.79\%$
test_ddpg_speed[reduce-overhead-backward] 2.3830ms 2.3365ms 427.9917 Ops/s 421.3697 Ops/s $\color{#35bf28}+1.57\%$
test_sac_speed[False-None] 8.4426ms 7.9087ms 126.4426 Ops/s 125.4776 Ops/s $\color{#35bf28}+0.77\%$
test_sac_speed[False-backward] 11.8973ms 11.1971ms 89.3088 Ops/s 89.2934 Ops/s $\color{#35bf28}+0.02\%$
test_sac_speed[True-None] 2.2306ms 2.0977ms 476.7175 Ops/s 474.8822 Ops/s $\color{#35bf28}+0.39\%$
test_sac_speed[True-backward] 4.1193ms 3.9944ms 250.3512 Ops/s 244.8100 Ops/s $\color{#35bf28}+2.26\%$
test_sac_speed[reduce-overhead-None] 2.2944ms 2.0830ms 480.0813 Ops/s 469.6263 Ops/s $\color{#35bf28}+2.23\%$
test_sac_speed[reduce-overhead-backward] 4.1265ms 4.0199ms 248.7629 Ops/s 229.9175 Ops/s $\textbf{\color{#35bf28}+8.20\%}$
test_redq_speed[False-None] 10.8562ms 10.4738ms 95.4761 Ops/s 95.5181 Ops/s $\color{#d91a1a}-0.04\%$
test_redq_speed[False-backward] 18.5218ms 17.8313ms 56.0812 Ops/s 56.0020 Ops/s $\color{#35bf28}+0.14\%$
test_redq_speed[True-None] 4.7631ms 4.3251ms 231.2067 Ops/s 222.4506 Ops/s $\color{#35bf28}+3.94\%$
test_redq_speed[True-backward] 10.3903ms 10.0066ms 99.9337 Ops/s 101.5713 Ops/s $\color{#d91a1a}-1.61\%$
test_redq_speed[reduce-overhead-None] 4.5166ms 4.3294ms 230.9775 Ops/s 229.2192 Ops/s $\color{#35bf28}+0.77\%$
test_redq_speed[reduce-overhead-backward] 10.5311ms 10.0846ms 99.1613 Ops/s 100.7822 Ops/s $\color{#d91a1a}-1.61\%$
test_redq_deprec_speed[False-None] 11.3194ms 10.9909ms 90.9841 Ops/s 91.3474 Ops/s $\color{#d91a1a}-0.40\%$
test_redq_deprec_speed[False-backward] 16.2503ms 15.7498ms 63.4929 Ops/s 64.2941 Ops/s $\color{#d91a1a}-1.25\%$
test_redq_deprec_speed[True-None] 4.0075ms 3.6731ms 272.2533 Ops/s 279.3826 Ops/s $\color{#d91a1a}-2.55\%$
test_redq_deprec_speed[True-backward] 7.9415ms 7.5878ms 131.7913 Ops/s 118.2952 Ops/s $\textbf{\color{#35bf28}+11.41\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.9242ms 3.6328ms 275.2682 Ops/s 273.6236 Ops/s $\color{#35bf28}+0.60\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.1089ms 7.6435ms 130.8293 Ops/s 133.6523 Ops/s $\color{#d91a1a}-2.11\%$
test_td3_speed[False-None] 8.5933ms 8.0200ms 124.6889 Ops/s 118.6660 Ops/s $\textbf{\color{#35bf28}+5.08\%}$
test_td3_speed[False-backward] 11.5584ms 10.9071ms 91.6836 Ops/s 90.9471 Ops/s $\color{#35bf28}+0.81\%$
test_td3_speed[True-None] 1.8114ms 1.7734ms 563.8981 Ops/s 552.9866 Ops/s $\color{#35bf28}+1.97\%$
test_td3_speed[True-backward] 3.7254ms 3.5684ms 280.2352 Ops/s 270.7082 Ops/s $\color{#35bf28}+3.52\%$
test_td3_speed[reduce-overhead-None] 1.7829ms 1.7431ms 573.7038 Ops/s 556.1732 Ops/s $\color{#35bf28}+3.15\%$
test_td3_speed[reduce-overhead-backward] 3.7292ms 3.6267ms 275.7364 Ops/s 242.0222 Ops/s $\textbf{\color{#35bf28}+13.93\%}$
test_cql_speed[False-None] 28.8253ms 26.0584ms 38.3753 Ops/s 38.5419 Ops/s $\color{#d91a1a}-0.43\%$
test_cql_speed[False-backward] 38.6426ms 35.6885ms 28.0203 Ops/s 28.6062 Ops/s $\color{#d91a1a}-2.05\%$
test_cql_speed[True-None] 12.8871ms 12.3395ms 81.0403 Ops/s 82.5347 Ops/s $\color{#d91a1a}-1.81\%$
test_cql_speed[True-backward] 18.7581ms 18.3932ms 54.3678 Ops/s 54.8520 Ops/s $\color{#d91a1a}-0.88\%$
test_cql_speed[reduce-overhead-None] 13.3199ms 12.4233ms 80.4939 Ops/s 82.1358 Ops/s $\color{#d91a1a}-2.00\%$
test_cql_speed[reduce-overhead-backward] 18.7447ms 18.3948ms 54.3631 Ops/s 53.7692 Ops/s $\color{#35bf28}+1.10\%$
test_a2c_speed[False-None] 5.6499ms 5.3213ms 187.9232 Ops/s 182.9426 Ops/s $\color{#35bf28}+2.72\%$
test_a2c_speed[False-backward] 12.2299ms 11.8311ms 84.5229 Ops/s 83.4712 Ops/s $\color{#35bf28}+1.26\%$
test_a2c_speed[True-None] 3.8424ms 3.6520ms 273.8192 Ops/s 259.7906 Ops/s $\textbf{\color{#35bf28}+5.40\%}$
test_a2c_speed[True-backward] 8.9443ms 8.6438ms 115.6900 Ops/s 114.9448 Ops/s $\color{#35bf28}+0.65\%$
test_a2c_speed[reduce-overhead-None] 3.8425ms 3.7319ms 267.9625 Ops/s 270.8754 Ops/s $\color{#d91a1a}-1.08\%$
test_a2c_speed[reduce-overhead-backward] 8.9705ms 8.8135ms 113.4628 Ops/s 112.5355 Ops/s $\color{#35bf28}+0.82\%$
test_ppo_speed[False-None] 6.3304ms 5.9269ms 168.7231 Ops/s 166.3508 Ops/s $\color{#35bf28}+1.43\%$
test_ppo_speed[False-backward] 12.9291ms 12.6603ms 78.9874 Ops/s 80.1020 Ops/s $\color{#d91a1a}-1.39\%$
test_ppo_speed[True-None] 3.9671ms 3.6554ms 273.5699 Ops/s 273.6004 Ops/s $\color{#d91a1a}-0.01\%$
test_ppo_speed[True-backward] 8.6235ms 8.4550ms 118.2730 Ops/s 104.9809 Ops/s $\textbf{\color{#35bf28}+12.66\%}$
test_ppo_speed[reduce-overhead-None] 3.9887ms 3.6358ms 275.0436 Ops/s 272.6816 Ops/s $\color{#35bf28}+0.87\%$
test_ppo_speed[reduce-overhead-backward] 9.0235ms 8.7258ms 114.6028 Ops/s 110.6042 Ops/s $\color{#35bf28}+3.62\%$
test_reinforce_speed[False-None] 4.9463ms 4.5959ms 217.5852 Ops/s 215.4820 Ops/s $\color{#35bf28}+0.98\%$
test_reinforce_speed[False-backward] 7.7725ms 7.4417ms 134.3775 Ops/s 131.3465 Ops/s $\color{#35bf28}+2.31\%$
test_reinforce_speed[True-None] 3.2214ms 2.8429ms 351.7574 Ops/s 342.0726 Ops/s $\color{#35bf28}+2.83\%$
test_reinforce_speed[True-backward] 8.0542ms 7.7631ms 128.8144 Ops/s 125.7185 Ops/s $\color{#35bf28}+2.46\%$
test_reinforce_speed[reduce-overhead-None] 3.1976ms 2.8468ms 351.2727 Ops/s 337.7669 Ops/s $\color{#35bf28}+4.00\%$
test_reinforce_speed[reduce-overhead-backward] 8.3301ms 7.8928ms 126.6977 Ops/s 115.9844 Ops/s $\textbf{\color{#35bf28}+9.24\%}$
test_iql_speed[False-None] 26.0569ms 20.3655ms 49.1026 Ops/s 48.6689 Ops/s $\color{#35bf28}+0.89\%$
test_iql_speed[False-backward] 35.7979ms 30.6893ms 32.5846 Ops/s 32.6787 Ops/s $\color{#d91a1a}-0.29\%$
test_iql_speed[True-None] 8.8629ms 8.4728ms 118.0251 Ops/s 117.7180 Ops/s $\color{#35bf28}+0.26\%$
test_iql_speed[True-backward] 17.3166ms 16.8305ms 59.4159 Ops/s 58.2054 Ops/s $\color{#35bf28}+2.08\%$
test_iql_speed[reduce-overhead-None] 8.8297ms 8.5479ms 116.9881 Ops/s 111.5563 Ops/s $\color{#35bf28}+4.87\%$
test_iql_speed[reduce-overhead-backward] 17.7438ms 17.1260ms 58.3907 Ops/s 57.4894 Ops/s $\color{#35bf28}+1.57\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.5635ms 6.0840ms 164.3650 Ops/s 165.2057 Ops/s $\color{#d91a1a}-0.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6092ms 0.3752ms 2.6653 KOps/s 3.2379 KOps/s $\textbf{\color{#d91a1a}-17.68\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6986ms 0.2601ms 3.8442 KOps/s 3.2663 KOps/s $\textbf{\color{#35bf28}+17.69\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0219ms 5.7813ms 172.9706 Ops/s 174.9987 Ops/s $\color{#d91a1a}-1.16\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8421ms 0.3213ms 3.1126 KOps/s 3.3143 KOps/s $\textbf{\color{#d91a1a}-6.09\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6151ms 0.3202ms 3.1235 KOps/s 3.3285 KOps/s $\textbf{\color{#d91a1a}-6.16\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7203ms 1.3199ms 757.6275 Ops/s 763.8493 Ops/s $\color{#d91a1a}-0.81\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5224ms 1.3051ms 766.2010 Ops/s 769.9813 Ops/s $\color{#d91a1a}-0.49\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1904ms 5.9498ms 168.0727 Ops/s 170.2872 Ops/s $\color{#d91a1a}-1.30\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9946ms 0.4812ms 2.0783 KOps/s 2.0334 KOps/s $\color{#35bf28}+2.21\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7354ms 0.4638ms 2.1561 KOps/s 2.0419 KOps/s $\textbf{\color{#35bf28}+5.59\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9710ms 5.8576ms 170.7188 Ops/s 174.3748 Ops/s $\color{#d91a1a}-2.10\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0759ms 0.3515ms 2.8448 KOps/s 730.7658 Ops/s $\textbf{\color{#35bf28}+289.29\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5052ms 0.3104ms 3.2218 KOps/s 3.7719 KOps/s $\textbf{\color{#d91a1a}-14.58\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1004ms 5.7453ms 174.0552 Ops/s 171.9927 Ops/s $\color{#35bf28}+1.20\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.8187ms 0.3155ms 3.1700 KOps/s 3.4870 KOps/s $\textbf{\color{#d91a1a}-9.09\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5055ms 0.2982ms 3.3539 KOps/s 3.4791 KOps/s $\color{#d91a1a}-3.60\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0539ms 5.9167ms 169.0124 Ops/s 166.8351 Ops/s $\color{#35bf28}+1.31\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0651ms 0.5150ms 1.9417 KOps/s 2.0481 KOps/s $\textbf{\color{#d91a1a}-5.19\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6965ms 0.4474ms 2.2349 KOps/s 2.1751 KOps/s $\color{#35bf28}+2.75\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5494s 15.9673ms 62.6281 Ops/s 195.3235 Ops/s $\textbf{\color{#d91a1a}-67.94\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.5803ms 2.0699ms 483.1150 Ops/s 427.9223 Ops/s $\textbf{\color{#35bf28}+12.90\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4159ms 1.1336ms 882.1673 Ops/s 825.1436 Ops/s $\textbf{\color{#35bf28}+6.91\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.5586ms 5.0384ms 198.4752 Ops/s 56.0513 Ops/s $\textbf{\color{#35bf28}+254.10\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 12.0566ms 1.9492ms 513.0363 Ops/s 479.7816 Ops/s $\textbf{\color{#35bf28}+6.93\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 3.3566ms 1.1015ms 907.8586 Ops/s 838.6350 Ops/s $\textbf{\color{#35bf28}+8.25\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.6790ms 5.2265ms 191.3315 Ops/s 185.9409 Ops/s $\color{#35bf28}+2.90\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.1097ms 2.1764ms 459.4763 Ops/s 446.8160 Ops/s $\color{#35bf28}+2.83\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0598ms 1.0923ms 915.4850 Ops/s 733.4381 Ops/s $\textbf{\color{#35bf28}+24.82\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.4701ms 32.9688ms 30.3317 Ops/s 29.9384 Ops/s $\color{#35bf28}+1.31\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.2947ms 17.7594ms 56.3083 Ops/s 55.5881 Ops/s $\color{#35bf28}+1.30\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 35.3571ms 33.4397ms 29.9045 Ops/s 29.1178 Ops/s $\color{#35bf28}+2.70\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.9026ms 18.0312ms 55.4593 Ops/s 54.9891 Ops/s $\color{#35bf28}+0.86\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 36.9746ms 35.4381ms 28.2182 Ops/s 27.8767 Ops/s $\color{#35bf28}+1.23\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 20.9459ms 19.2831ms 51.8588 Ops/s 50.6973 Ops/s $\color{#35bf28}+2.29\%$

[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
vmoens added 14 commits October 22, 2025 12:31
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
[ghstack-poisoned]
@vmoens vmoens changed the title [Feature] SAC Trainer [Feature] Add timing options Oct 25, 2025
vmoens added a commit that referenced this pull request Oct 25, 2025
ghstack-source-id: 15bba74
Pull-Request: #3191
@vmoens vmoens merged commit 7a28716 into gh/vmoens/154/base Oct 25, 2025
88 of 101 checks passed
@vmoens vmoens deleted the gh/vmoens/154/head branch October 25, 2025 16:39
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.

1 participant