Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Dec 31, 2025

[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/3286

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:

❌ 5 New Failures

As of commit d11a914 with merge base 7866d11 (image):

NEW FAILURES - The following jobs have failed:

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

vmoens added a commit that referenced this pull request Dec 31, 2025
ghstack-source-id: 2e1839f
Pull-Request: #3286
@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 Dec 31, 2025
@github-actions
Copy link

github-actions bot commented Dec 31, 2025

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

Total Benchmarks: 164. Improved: $\large\color{#35bf28}24$. 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] 82.3779μs 81.3792μs 12.2882 KOps/s 11.5353 KOps/s $\textbf{\color{#35bf28}+6.53\%}$
test_tensor_to_bytestream_speed[torch.save] 0.1414ms 0.1406ms 7.1106 KOps/s 6.9450 KOps/s $\color{#35bf28}+2.38\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1192s 0.1189s 8.4080 Ops/s 7.6109 Ops/s $\textbf{\color{#35bf28}+10.47\%}$
test_tensor_to_bytestream_speed[numpy] 2.7312μs 2.7242μs 367.0755 KOps/s 360.1137 KOps/s $\color{#35bf28}+1.93\%$
test_tensor_to_bytestream_speed[safetensors] 38.6893μs 38.5026μs 25.9723 KOps/s 25.9564 KOps/s $\color{#35bf28}+0.06\%$
test_simple 0.5648s 0.5615s 1.7809 Ops/s 1.7044 Ops/s $\color{#35bf28}+4.49\%$
test_transformed 1.1463s 1.1444s 0.8739 Ops/s 0.8570 Ops/s $\color{#35bf28}+1.97\%$
test_serial 1.7173s 1.7102s 0.5847 Ops/s 0.5813 Ops/s $\color{#35bf28}+0.58\%$
test_parallel 1.0987s 1.0949s 0.9133 Ops/s 0.8774 Ops/s $\color{#35bf28}+4.09\%$
test_step_mdp_speed[True-True-True-True-True] 0.3279ms 45.1325μs 22.1570 KOps/s 22.6519 KOps/s $\color{#d91a1a}-2.18\%$
test_step_mdp_speed[True-True-True-True-False] 54.3210μs 25.2614μs 39.5860 KOps/s 39.6385 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-True-True-False-True] 77.7020μs 25.8064μs 38.7501 KOps/s 38.8764 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[True-True-True-False-False] 41.5900μs 14.2894μs 69.9821 KOps/s 70.9345 KOps/s $\color{#d91a1a}-1.34\%$
test_step_mdp_speed[True-True-False-True-True] 95.4320μs 49.8185μs 20.0729 KOps/s 20.8825 KOps/s $\color{#d91a1a}-3.88\%$
test_step_mdp_speed[True-True-False-True-False] 73.8810μs 28.2916μs 35.3461 KOps/s 35.3080 KOps/s $\color{#35bf28}+0.11\%$
test_step_mdp_speed[True-True-False-False-True] 54.4010μs 28.3218μs 35.3085 KOps/s 35.5932 KOps/s $\color{#d91a1a}-0.80\%$
test_step_mdp_speed[True-True-False-False-False] 52.9110μs 17.0829μs 58.5380 KOps/s 58.6355 KOps/s $\color{#d91a1a}-0.17\%$
test_step_mdp_speed[True-False-True-True-True] 83.0110μs 51.8208μs 19.2973 KOps/s 19.6413 KOps/s $\color{#d91a1a}-1.75\%$
test_step_mdp_speed[True-False-True-True-False] 68.1110μs 31.2801μs 31.9692 KOps/s 32.4237 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[True-False-True-False-True] 58.0310μs 28.2157μs 35.4412 KOps/s 35.2824 KOps/s $\color{#35bf28}+0.45\%$
test_step_mdp_speed[True-False-True-False-False] 45.9510μs 17.1122μs 58.4377 KOps/s 59.0984 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[True-False-False-True-True] 90.3120μs 55.3495μs 18.0670 KOps/s 18.7512 KOps/s $\color{#d91a1a}-3.65\%$
test_step_mdp_speed[True-False-False-True-False] 66.2710μs 35.5975μs 28.0919 KOps/s 29.5466 KOps/s $\color{#d91a1a}-4.92\%$
test_step_mdp_speed[True-False-False-False-True] 64.5710μs 30.7225μs 32.5494 KOps/s 33.1254 KOps/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[True-False-False-False-False] 56.1400μs 19.7311μs 50.6815 KOps/s 50.9309 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[False-True-True-True-True] 80.3110μs 51.6279μs 19.3694 KOps/s 19.2083 KOps/s $\color{#35bf28}+0.84\%$
test_step_mdp_speed[False-True-True-True-False] 64.4610μs 31.5747μs 31.6709 KOps/s 32.2399 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[False-True-True-False-True] 2.3042ms 33.2608μs 30.0654 KOps/s 31.0473 KOps/s $\color{#d91a1a}-3.16\%$
test_step_mdp_speed[False-True-True-False-False] 48.5110μs 18.9001μs 52.9098 KOps/s 53.8548 KOps/s $\color{#d91a1a}-1.75\%$
test_step_mdp_speed[False-True-False-True-True] 0.1346ms 56.4246μs 17.7228 KOps/s 18.8216 KOps/s $\textbf{\color{#d91a1a}-5.84\%}$
test_step_mdp_speed[False-True-False-True-False] 92.2810μs 35.4336μs 28.2218 KOps/s 29.5115 KOps/s $\color{#d91a1a}-4.37\%$
test_step_mdp_speed[False-True-False-False-True] 90.1420μs 35.2394μs 28.3773 KOps/s 29.0759 KOps/s $\color{#d91a1a}-2.40\%$
test_step_mdp_speed[False-True-False-False-False] 48.9210μs 21.9527μs 45.5524 KOps/s 47.1722 KOps/s $\color{#d91a1a}-3.43\%$
test_step_mdp_speed[False-False-True-True-True] 95.1010μs 57.3794μs 17.4279 KOps/s 17.8458 KOps/s $\color{#d91a1a}-2.34\%$
test_step_mdp_speed[False-False-True-True-False] 0.1128ms 36.9596μs 27.0566 KOps/s 27.5152 KOps/s $\color{#d91a1a}-1.67\%$
test_step_mdp_speed[False-False-True-False-True] 70.1910μs 35.0855μs 28.5018 KOps/s 29.1777 KOps/s $\color{#d91a1a}-2.32\%$
test_step_mdp_speed[False-False-True-False-False] 48.7900μs 21.7167μs 46.0476 KOps/s 47.4520 KOps/s $\color{#d91a1a}-2.96\%$
test_step_mdp_speed[False-False-False-True-True] 0.1021ms 59.2968μs 16.8643 KOps/s 17.1920 KOps/s $\color{#d91a1a}-1.91\%$
test_step_mdp_speed[False-False-False-True-False] 75.3110μs 39.2487μs 25.4786 KOps/s 25.5670 KOps/s $\color{#d91a1a}-0.35\%$
test_step_mdp_speed[False-False-False-False-True] 0.1207ms 37.2009μs 26.8811 KOps/s 27.1473 KOps/s $\color{#d91a1a}-0.98\%$
test_step_mdp_speed[False-False-False-False-False] 60.1810μs 24.2886μs 41.1715 KOps/s 42.1396 KOps/s $\color{#d91a1a}-2.30\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8887s 0.7960s 1.2562 Ops/s 1.2638 Ops/s $\color{#d91a1a}-0.60\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7460s 0.6476s 1.5443 Ops/s 1.5289 Ops/s $\color{#35bf28}+1.01\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7954s 1.7220s 0.5807 Ops/s 0.5802 Ops/s $\color{#35bf28}+0.10\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5622s 1.4835s 0.6741 Ops/s 0.6682 Ops/s $\color{#35bf28}+0.88\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0425s 1.9683s 0.5080 Ops/s 0.5071 Ops/s $\color{#35bf28}+0.19\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.8331s 1.7543s 0.5700 Ops/s 0.5718 Ops/s $\color{#d91a1a}-0.31\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8446s 4.7525s 0.2104 Ops/s 0.2134 Ops/s $\color{#d91a1a}-1.41\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6434s 4.5221s 0.2211 Ops/s 0.2183 Ops/s $\color{#35bf28}+1.31\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1168s 2.0353s 0.4913 Ops/s 0.4737 Ops/s $\color{#35bf28}+3.71\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7792s 1.6902s 0.5916 Ops/s 0.5870 Ops/s $\color{#35bf28}+0.80\%$
test_values[generalized_advantage_estimate-True-True] 11.4205ms 10.6357ms 94.0231 Ops/s 95.4589 Ops/s $\color{#d91a1a}-1.50\%$
test_values[vec_generalized_advantage_estimate-True-True] 12.9929ms 11.0882ms 90.1859 Ops/s 89.6117 Ops/s $\color{#35bf28}+0.64\%$
test_values[td0_return_estimate-False-False] 0.2622ms 0.1313ms 7.6168 KOps/s 7.4742 KOps/s $\color{#35bf28}+1.91\%$
test_values[td1_return_estimate-False-False] 28.4132ms 27.9620ms 35.7628 Ops/s 35.8963 Ops/s $\color{#d91a1a}-0.37\%$
test_values[vec_td1_return_estimate-False-False] 17.6752ms 11.7135ms 85.3717 Ops/s 89.4060 Ops/s $\color{#d91a1a}-4.51\%$
test_values[td_lambda_return_estimate-True-False] 44.5258ms 41.5147ms 24.0879 Ops/s 23.8299 Ops/s $\color{#35bf28}+1.08\%$
test_values[vec_td_lambda_return_estimate-True-False] 17.6774ms 11.5043ms 86.9237 Ops/s 88.7709 Ops/s $\color{#d91a1a}-2.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.4154ms 9.3385ms 107.0834 Ops/s 106.7344 Ops/s $\color{#35bf28}+0.33\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7579ms 1.4705ms 680.0249 Ops/s 658.3727 Ops/s $\color{#35bf28}+3.29\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4714ms 0.4199ms 2.3813 KOps/s 2.3010 KOps/s $\color{#35bf28}+3.49\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.0303ms 29.3496ms 34.0720 Ops/s 33.7245 Ops/s $\color{#35bf28}+1.03\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.0983ms 1.7229ms 580.4035 Ops/s 576.9334 Ops/s $\color{#35bf28}+0.60\%$
test_dqn_speed[False-None] 2.0751ms 1.4516ms 688.9170 Ops/s 696.5145 Ops/s $\color{#d91a1a}-1.09\%$
test_dqn_speed[False-backward] 1.9883ms 1.9451ms 514.1208 Ops/s 511.5902 Ops/s $\color{#35bf28}+0.49\%$
test_dqn_speed[True-None] 0.6963ms 0.5411ms 1.8482 KOps/s 1.7573 KOps/s $\textbf{\color{#35bf28}+5.17\%}$
test_dqn_speed[True-backward] 1.0550ms 0.9881ms 1.0120 KOps/s 1.0031 KOps/s $\color{#35bf28}+0.88\%$
test_dqn_speed[reduce-overhead-None] 0.7245ms 0.5337ms 1.8739 KOps/s 1.8768 KOps/s $\color{#d91a1a}-0.16\%$
test_dqn_speed[reduce-overhead-backward] 1.0355ms 0.9780ms 1.0225 KOps/s 865.8338 Ops/s $\textbf{\color{#35bf28}+18.09\%}$
test_ddpg_speed[False-None] 3.2880ms 2.9478ms 339.2349 Ops/s 330.4374 Ops/s $\color{#35bf28}+2.66\%$
test_ddpg_speed[False-backward] 4.5827ms 4.1773ms 239.3871 Ops/s 240.9491 Ops/s $\color{#d91a1a}-0.65\%$
test_ddpg_speed[True-None] 1.8016ms 1.4026ms 712.9459 Ops/s 712.9153 Ops/s $+0.00\%$
test_ddpg_speed[True-backward] 2.8414ms 2.4311ms 411.3323 Ops/s 360.3237 Ops/s $\textbf{\color{#35bf28}+14.16\%}$
test_ddpg_speed[reduce-overhead-None] 1.7869ms 1.4062ms 711.1341 Ops/s 714.8396 Ops/s $\color{#d91a1a}-0.52\%$
test_ddpg_speed[reduce-overhead-backward] 2.4666ms 2.3684ms 422.2244 Ops/s 385.0303 Ops/s $\textbf{\color{#35bf28}+9.66\%}$
test_sac_speed[False-None] 8.5729ms 8.0900ms 123.6095 Ops/s 124.3445 Ops/s $\color{#d91a1a}-0.59\%$
test_sac_speed[False-backward] 11.7881ms 11.4516ms 87.3237 Ops/s 88.6220 Ops/s $\color{#d91a1a}-1.46\%$
test_sac_speed[True-None] 2.5880ms 2.1494ms 465.2355 Ops/s 459.3742 Ops/s $\color{#35bf28}+1.28\%$
test_sac_speed[True-backward] 6.5704ms 4.3190ms 231.5342 Ops/s 248.3260 Ops/s $\textbf{\color{#d91a1a}-6.76\%}$
test_sac_speed[reduce-overhead-None] 2.2694ms 2.1300ms 469.4902 Ops/s 455.5877 Ops/s $\color{#35bf28}+3.05\%$
test_sac_speed[reduce-overhead-backward] 4.4264ms 4.0644ms 246.0376 Ops/s 232.4411 Ops/s $\textbf{\color{#35bf28}+5.85\%}$
test_redq_speed[False-None] 11.4574ms 10.5144ms 95.1074 Ops/s 95.6447 Ops/s $\color{#d91a1a}-0.56\%$
test_redq_speed[False-backward] 19.1894ms 18.1698ms 55.0365 Ops/s 55.2998 Ops/s $\color{#d91a1a}-0.48\%$
test_redq_speed[True-None] 4.7722ms 4.4507ms 224.6819 Ops/s 226.7687 Ops/s $\color{#d91a1a}-0.92\%$
test_redq_speed[True-backward] 10.0279ms 9.7793ms 102.2569 Ops/s 106.8692 Ops/s $\color{#d91a1a}-4.32\%$
test_redq_speed[reduce-overhead-None] 6.1943ms 4.5388ms 220.3232 Ops/s 228.5536 Ops/s $\color{#d91a1a}-3.60\%$
test_redq_speed[reduce-overhead-backward] 10.2346ms 9.9204ms 100.8029 Ops/s 104.8804 Ops/s $\color{#d91a1a}-3.89\%$
test_redq_deprec_speed[False-None] 11.8028ms 11.2608ms 88.8036 Ops/s 91.7039 Ops/s $\color{#d91a1a}-3.16\%$
test_redq_deprec_speed[False-backward] 16.8496ms 16.3433ms 61.1871 Ops/s 64.2040 Ops/s $\color{#d91a1a}-4.70\%$
test_redq_deprec_speed[True-None] 3.9910ms 3.6906ms 270.9616 Ops/s 274.3193 Ops/s $\color{#d91a1a}-1.22\%$
test_redq_deprec_speed[True-backward] 7.8985ms 7.6388ms 130.9100 Ops/s 124.5740 Ops/s $\textbf{\color{#35bf28}+5.09\%}$
test_redq_deprec_speed[reduce-overhead-None] 4.0820ms 3.6528ms 273.7644 Ops/s 262.4866 Ops/s $\color{#35bf28}+4.30\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.8539ms 7.6211ms 131.2153 Ops/s 124.7306 Ops/s $\textbf{\color{#35bf28}+5.20\%}$
test_td3_speed[False-None] 9.0865ms 8.2688ms 120.9362 Ops/s 120.8975 Ops/s $\color{#35bf28}+0.03\%$
test_td3_speed[False-backward] 11.6490ms 11.0890ms 90.1798 Ops/s 89.8213 Ops/s $\color{#35bf28}+0.40\%$
test_td3_speed[True-None] 1.8776ms 1.8405ms 543.3432 Ops/s 535.2918 Ops/s $\color{#35bf28}+1.50\%$
test_td3_speed[True-backward] 3.7523ms 3.6177ms 276.4163 Ops/s 240.4028 Ops/s $\textbf{\color{#35bf28}+14.98\%}$
test_td3_speed[reduce-overhead-None] 1.8419ms 1.8059ms 553.7520 Ops/s 533.1560 Ops/s $\color{#35bf28}+3.86\%$
test_td3_speed[reduce-overhead-backward] 3.8295ms 3.6529ms 273.7519 Ops/s 223.4448 Ops/s $\textbf{\color{#35bf28}+22.51\%}$
test_cql_speed[False-None] 26.8563ms 26.0427ms 38.3984 Ops/s 37.9794 Ops/s $\color{#35bf28}+1.10\%$
test_cql_speed[False-backward] 36.3678ms 35.5775ms 28.1076 Ops/s 27.8205 Ops/s $\color{#35bf28}+1.03\%$
test_cql_speed[True-None] 12.9124ms 12.4221ms 80.5018 Ops/s 81.8723 Ops/s $\color{#d91a1a}-1.67\%$
test_cql_speed[True-backward] 19.0030ms 18.4054ms 54.3318 Ops/s 55.4339 Ops/s $\color{#d91a1a}-1.99\%$
test_cql_speed[reduce-overhead-None] 13.1308ms 12.7026ms 78.7243 Ops/s 79.5644 Ops/s $\color{#d91a1a}-1.06\%$
test_cql_speed[reduce-overhead-backward] 19.4223ms 18.4566ms 54.1813 Ops/s 56.0335 Ops/s $\color{#d91a1a}-3.31\%$
test_a2c_speed[False-None] 5.8658ms 5.4929ms 182.0530 Ops/s 180.3100 Ops/s $\color{#35bf28}+0.97\%$
test_a2c_speed[False-backward] 12.3372ms 11.9192ms 83.8984 Ops/s 82.8574 Ops/s $\color{#35bf28}+1.26\%$
test_a2c_speed[True-None] 4.0386ms 3.6988ms 270.3557 Ops/s 265.8468 Ops/s $\color{#35bf28}+1.70\%$
test_a2c_speed[True-backward] 9.3074ms 8.6858ms 115.1305 Ops/s 105.5991 Ops/s $\textbf{\color{#35bf28}+9.03\%}$
test_a2c_speed[reduce-overhead-None] 3.8450ms 3.7039ms 269.9834 Ops/s 266.5133 Ops/s $\color{#35bf28}+1.30\%$
test_a2c_speed[reduce-overhead-backward] 8.8629ms 8.6753ms 115.2692 Ops/s 109.5164 Ops/s $\textbf{\color{#35bf28}+5.25\%}$
test_ppo_speed[False-None] 6.3283ms 5.9643ms 167.6642 Ops/s 166.1901 Ops/s $\color{#35bf28}+0.89\%$
test_ppo_speed[False-backward] 13.0698ms 12.6305ms 79.1737 Ops/s 79.5153 Ops/s $\color{#d91a1a}-0.43\%$
test_ppo_speed[True-None] 3.8123ms 3.6452ms 274.3312 Ops/s 268.4165 Ops/s $\color{#35bf28}+2.20\%$
test_ppo_speed[True-backward] 8.6864ms 8.3957ms 119.1080 Ops/s 112.7698 Ops/s $\textbf{\color{#35bf28}+5.62\%}$
test_ppo_speed[reduce-overhead-None] 3.7678ms 3.6058ms 277.3292 Ops/s 275.8727 Ops/s $\color{#35bf28}+0.53\%$
test_ppo_speed[reduce-overhead-backward] 8.9597ms 8.7172ms 114.7161 Ops/s 113.6043 Ops/s $\color{#35bf28}+0.98\%$
test_reinforce_speed[False-None] 5.1090ms 4.5941ms 217.6702 Ops/s 217.5958 Ops/s $\color{#35bf28}+0.03\%$
test_reinforce_speed[False-backward] 7.7378ms 7.4962ms 133.4002 Ops/s 135.2225 Ops/s $\color{#d91a1a}-1.35\%$
test_reinforce_speed[True-None] 10.1132ms 2.9916ms 334.2702 Ops/s 343.2140 Ops/s $\color{#d91a1a}-2.61\%$
test_reinforce_speed[True-backward] 10.3546ms 8.1922ms 122.0669 Ops/s 127.2444 Ops/s $\color{#d91a1a}-4.07\%$
test_reinforce_speed[reduce-overhead-None] 3.2275ms 2.8672ms 348.7776 Ops/s 343.0233 Ops/s $\color{#35bf28}+1.68\%$
test_reinforce_speed[reduce-overhead-backward] 8.1291ms 7.8880ms 126.7747 Ops/s 124.4366 Ops/s $\color{#35bf28}+1.88\%$
test_iql_speed[False-None] 24.4297ms 20.0604ms 49.8494 Ops/s 48.3917 Ops/s $\color{#35bf28}+3.01\%$
test_iql_speed[False-backward] 31.4890ms 30.5696ms 32.7123 Ops/s 32.3614 Ops/s $\color{#35bf28}+1.08\%$
test_iql_speed[True-None] 8.7766ms 8.4989ms 117.6624 Ops/s 114.6182 Ops/s $\color{#35bf28}+2.66\%$
test_iql_speed[True-backward] 17.1449ms 16.6386ms 60.1012 Ops/s 59.2295 Ops/s $\color{#35bf28}+1.47\%$
test_iql_speed[reduce-overhead-None] 9.0316ms 8.5765ms 116.5974 Ops/s 114.7232 Ops/s $\color{#35bf28}+1.63\%$
test_iql_speed[reduce-overhead-backward] 17.7657ms 17.2092ms 58.1086 Ops/s 57.8275 Ops/s $\color{#35bf28}+0.49\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.9986ms 6.1118ms 163.6191 Ops/s 161.7856 Ops/s $\color{#35bf28}+1.13\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6102ms 0.3582ms 2.7918 KOps/s 2.5713 KOps/s $\textbf{\color{#35bf28}+8.58\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6489ms 0.3859ms 2.5915 KOps/s 2.6817 KOps/s $\color{#d91a1a}-3.36\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0437ms 5.8118ms 172.0636 Ops/s 170.3954 Ops/s $\color{#35bf28}+0.98\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7099ms 0.3610ms 2.7698 KOps/s 2.6309 KOps/s $\textbf{\color{#35bf28}+5.28\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6409ms 0.3774ms 2.6496 KOps/s 3.2183 KOps/s $\textbf{\color{#d91a1a}-17.67\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7503ms 1.4640ms 683.0517 Ops/s 699.9049 Ops/s $\color{#d91a1a}-2.41\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5926ms 1.3800ms 724.6485 Ops/s 745.8878 Ops/s $\color{#d91a1a}-2.85\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.0039ms 6.1051ms 163.7979 Ops/s 165.3842 Ops/s $\color{#d91a1a}-0.96\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8455ms 0.4397ms 2.2745 KOps/s 1.9175 KOps/s $\textbf{\color{#35bf28}+18.62\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7071ms 0.4373ms 2.2870 KOps/s 1.9875 KOps/s $\textbf{\color{#35bf28}+15.07\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9061ms 5.8314ms 171.4856 Ops/s 168.0480 Ops/s $\color{#35bf28}+2.05\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.5808ms 0.3222ms 3.1039 KOps/s 3.1141 KOps/s $\color{#d91a1a}-0.33\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4548ms 0.2699ms 3.7045 KOps/s 3.3195 KOps/s $\textbf{\color{#35bf28}+11.60\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0417ms 5.6911ms 175.7121 Ops/s 170.1590 Ops/s $\color{#35bf28}+3.26\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9010ms 0.3568ms 2.8026 KOps/s 3.3808 KOps/s $\textbf{\color{#d91a1a}-17.10\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5079ms 0.3430ms 2.9158 KOps/s 3.2876 KOps/s $\textbf{\color{#d91a1a}-11.31\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0186ms 5.9233ms 168.8238 Ops/s 165.7282 Ops/s $\color{#35bf28}+1.87\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0975ms 0.4905ms 2.0388 KOps/s 1.8706 KOps/s $\textbf{\color{#35bf28}+8.99\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7745ms 0.4877ms 2.0505 KOps/s 1.9495 KOps/s $\textbf{\color{#35bf28}+5.19\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.5990ms 5.0058ms 199.7696 Ops/s 190.7858 Ops/s $\color{#35bf28}+4.71\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.9200ms 2.3505ms 425.4500 Ops/s 417.2376 Ops/s $\color{#35bf28}+1.97\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.7330ms 1.2063ms 828.9923 Ops/s 818.9977 Ops/s $\color{#35bf28}+1.22\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.4381ms 5.0291ms 198.8431 Ops/s 52.5946 Ops/s $\textbf{\color{#35bf28}+278.07\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.5610ms 2.3545ms 424.7171 Ops/s 500.3967 Ops/s $\textbf{\color{#d91a1a}-15.12\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.6069s 13.2895ms 75.2472 Ops/s 805.5243 Ops/s $\textbf{\color{#d91a1a}-90.66\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.6338ms 5.2590ms 190.1515 Ops/s 184.0294 Ops/s $\color{#35bf28}+3.33\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.4850ms 2.2655ms 441.4089 Ops/s 428.3879 Ops/s $\color{#35bf28}+3.04\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0218ms 1.0744ms 930.7612 Ops/s 737.9587 Ops/s $\textbf{\color{#35bf28}+26.13\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 38.5601ms 34.3911ms 29.0773 Ops/s 28.8684 Ops/s $\color{#35bf28}+0.72\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.0143ms 18.1113ms 55.2141 Ops/s 55.8025 Ops/s $\color{#d91a1a}-1.05\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 37.3988ms 34.9410ms 28.6196 Ops/s 27.1913 Ops/s $\textbf{\color{#35bf28}+5.25\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.7458ms 18.8394ms 53.0803 Ops/s 54.5603 Ops/s $\color{#d91a1a}-2.71\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 42.1859ms 37.0692ms 26.9766 Ops/s 26.1220 Ops/s $\color{#35bf28}+3.27\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.5831ms 20.2870ms 49.2926 Ops/s 48.7354 Ops/s $\color{#35bf28}+1.14\%$

@vmoens vmoens added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Dec 31, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 31, 2025
ghstack-source-id: 7c92a6d
Pull-Request: #3286
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 31, 2025
ghstack-source-id: d5fae60
Pull-Request: #3286
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