Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Dec 18, 2025

[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Dec 18, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3264

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

❌ 6 New Failures, 1 Cancelled Job, 1 Unrelated Failure

As of commit 00a7b9d with merge base 546a1b7 (image):

NEW FAILURES - The following jobs have failed:

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

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.

vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: 8629437
Pull-Request: #3264
@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 18, 2025
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: 57c33dc
Pull-Request: #3264
@github-actions
Copy link

github-actions bot commented Dec 18, 2025

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

Total Benchmarks: 164. Improved: $\large\color{#35bf28}18$. Worsened: $\large\color{#d91a1a}17$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.4740μs 80.9852μs 12.3479 KOps/s 12.2682 KOps/s $\color{#35bf28}+0.65\%$
test_tensor_to_bytestream_speed[torch.save] 0.1445ms 0.1439ms 6.9488 KOps/s 7.1311 KOps/s $\color{#d91a1a}-2.56\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1143s 0.1131s 8.8394 Ops/s 8.1065 Ops/s $\textbf{\color{#35bf28}+9.04\%}$
test_tensor_to_bytestream_speed[numpy] 2.6866μs 2.6673μs 374.9180 KOps/s 340.7271 KOps/s $\textbf{\color{#35bf28}+10.03\%}$
test_tensor_to_bytestream_speed[safetensors] 37.5062μs 37.1986μs 26.8827 KOps/s 25.9512 KOps/s $\color{#35bf28}+3.59\%$
test_simple 0.5482s 0.5457s 1.8323 Ops/s 1.7327 Ops/s $\textbf{\color{#35bf28}+5.75\%}$
test_transformed 1.1268s 1.1217s 0.8915 Ops/s 0.8585 Ops/s $\color{#35bf28}+3.84\%$
test_serial 1.6420s 1.6379s 0.6106 Ops/s 0.5896 Ops/s $\color{#35bf28}+3.56\%$
test_parallel 1.2233s 1.1741s 0.8517 Ops/s 0.8431 Ops/s $\color{#35bf28}+1.02\%$
test_step_mdp_speed[True-True-True-True-True] 0.2705ms 44.8854μs 22.2790 KOps/s 23.2848 KOps/s $\color{#d91a1a}-4.32\%$
test_step_mdp_speed[True-True-True-True-False] 76.8720μs 24.8432μs 40.2525 KOps/s 40.2227 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[True-True-True-False-True] 56.2720μs 25.2230μs 39.6463 KOps/s 41.0126 KOps/s $\color{#d91a1a}-3.33\%$
test_step_mdp_speed[True-True-True-False-False] 46.5210μs 13.8251μs 72.3324 KOps/s 73.5971 KOps/s $\color{#d91a1a}-1.72\%$
test_step_mdp_speed[True-True-False-True-True] 89.6920μs 47.4285μs 21.0844 KOps/s 21.2304 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[True-True-False-True-False] 56.9210μs 27.6252μs 36.1988 KOps/s 36.3279 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[True-True-False-False-True] 56.6410μs 27.6970μs 36.1051 KOps/s 36.6346 KOps/s $\color{#d91a1a}-1.45\%$
test_step_mdp_speed[True-True-False-False-False] 45.9410μs 16.3100μs 61.3121 KOps/s 61.4448 KOps/s $\color{#d91a1a}-0.22\%$
test_step_mdp_speed[True-False-True-True-True] 99.7520μs 50.2248μs 19.9105 KOps/s 20.0175 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[True-False-True-True-False] 66.7920μs 30.5141μs 32.7718 KOps/s 32.9516 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[True-False-True-False-True] 67.0810μs 27.5599μs 36.2846 KOps/s 36.6962 KOps/s $\color{#d91a1a}-1.12\%$
test_step_mdp_speed[True-False-True-False-False] 48.0610μs 16.2983μs 61.3561 KOps/s 60.9523 KOps/s $\color{#35bf28}+0.66\%$
test_step_mdp_speed[True-False-False-True-True] 0.1023ms 52.7262μs 18.9659 KOps/s 19.3205 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[True-False-False-True-False] 77.1820μs 32.5140μs 30.7560 KOps/s 30.2272 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[True-False-False-False-True] 74.2120μs 30.4314μs 32.8607 KOps/s 33.9342 KOps/s $\color{#d91a1a}-3.16\%$
test_step_mdp_speed[True-False-False-False-False] 52.2110μs 18.8040μs 53.1803 KOps/s 52.5979 KOps/s $\color{#35bf28}+1.11\%$
test_step_mdp_speed[False-True-True-True-True] 99.7830μs 50.3308μs 19.8686 KOps/s 20.0715 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-True-True-True-False] 57.1610μs 30.5452μs 32.7384 KOps/s 32.9847 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-True-True-False-True] 2.3164ms 31.5535μs 31.6922 KOps/s 31.6006 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[False-True-True-False-False] 48.8110μs 18.3795μs 54.4084 KOps/s 55.3422 KOps/s $\color{#d91a1a}-1.69\%$
test_step_mdp_speed[False-True-False-True-True] 87.4320μs 52.1022μs 19.1930 KOps/s 19.0087 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[False-True-False-True-False] 66.4720μs 33.5407μs 29.8146 KOps/s 30.1650 KOps/s $\color{#d91a1a}-1.16\%$
test_step_mdp_speed[False-True-False-False-True] 68.3820μs 34.0951μs 29.3297 KOps/s 29.5473 KOps/s $\color{#d91a1a}-0.74\%$
test_step_mdp_speed[False-True-False-False-False] 54.7510μs 21.0094μs 47.5977 KOps/s 47.9010 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[False-False-True-True-True] 88.0620μs 54.4455μs 18.3670 KOps/s 18.2790 KOps/s $\color{#35bf28}+0.48\%$
test_step_mdp_speed[False-False-True-True-False] 73.8510μs 35.6866μs 28.0217 KOps/s 27.8761 KOps/s $\color{#35bf28}+0.52\%$
test_step_mdp_speed[False-False-True-False-True] 63.3820μs 33.9574μs 29.4487 KOps/s 29.5282 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[False-False-True-False-False] 55.6710μs 20.8583μs 47.9426 KOps/s 48.2647 KOps/s $\color{#d91a1a}-0.67\%$
test_step_mdp_speed[False-False-False-True-True] 85.9720μs 56.9832μs 17.5490 KOps/s 17.5868 KOps/s $\color{#d91a1a}-0.21\%$
test_step_mdp_speed[False-False-False-True-False] 74.8120μs 38.4101μs 26.0348 KOps/s 26.1444 KOps/s $\color{#d91a1a}-0.42\%$
test_step_mdp_speed[False-False-False-False-True] 70.5720μs 36.3260μs 27.5285 KOps/s 28.2730 KOps/s $\color{#d91a1a}-2.63\%$
test_step_mdp_speed[False-False-False-False-False] 55.7410μs 22.9709μs 43.5333 KOps/s 42.6677 KOps/s $\color{#35bf28}+2.03\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8793s 0.7779s 1.2854 Ops/s 1.3082 Ops/s $\color{#d91a1a}-1.74\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7421s 0.6409s 1.5604 Ops/s 1.5839 Ops/s $\color{#d91a1a}-1.48\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7781s 1.6992s 0.5885 Ops/s 0.6004 Ops/s $\color{#d91a1a}-1.98\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5490s 1.4730s 0.6789 Ops/s 0.6932 Ops/s $\color{#d91a1a}-2.07\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0252s 1.9456s 0.5140 Ops/s 0.5231 Ops/s $\color{#d91a1a}-1.74\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.8036s 1.7180s 0.5821 Ops/s 0.5912 Ops/s $\color{#d91a1a}-1.54\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.7054s 4.5065s 0.2219 Ops/s 0.2216 Ops/s $\color{#35bf28}+0.16\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6490s 4.4659s 0.2239 Ops/s 0.2243 Ops/s $\color{#d91a1a}-0.15\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.1196s 1.9736s 0.5067 Ops/s 0.5050 Ops/s $\color{#35bf28}+0.34\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7819s 1.6715s 0.5983 Ops/s 0.5904 Ops/s $\color{#35bf28}+1.33\%$
test_values[generalized_advantage_estimate-True-True] 10.0113ms 9.5782ms 104.4033 Ops/s 97.6469 Ops/s $\textbf{\color{#35bf28}+6.92\%}$
test_values[vec_generalized_advantage_estimate-True-True] 13.6675ms 11.1173ms 89.9496 Ops/s 89.4736 Ops/s $\color{#35bf28}+0.53\%$
test_values[td0_return_estimate-False-False] 0.2290ms 0.1271ms 7.8679 KOps/s 7.8342 KOps/s $\color{#35bf28}+0.43\%$
test_values[td1_return_estimate-False-False] 25.9086ms 25.5184ms 39.1874 Ops/s 36.6317 Ops/s $\textbf{\color{#35bf28}+6.98\%}$
test_values[vec_td1_return_estimate-False-False] 11.9622ms 11.1993ms 89.2911 Ops/s 89.2726 Ops/s $\color{#35bf28}+0.02\%$
test_values[td_lambda_return_estimate-True-False] 40.0176ms 37.8820ms 26.3977 Ops/s 24.0863 Ops/s $\textbf{\color{#35bf28}+9.60\%}$
test_values[vec_td_lambda_return_estimate-True-False] 12.2237ms 11.1946ms 89.3284 Ops/s 90.2040 Ops/s $\color{#d91a1a}-0.97\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.7097ms 8.5334ms 117.1860 Ops/s 108.6240 Ops/s $\textbf{\color{#35bf28}+7.88\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.7396ms 1.5360ms 651.0205 Ops/s 663.8109 Ops/s $\color{#d91a1a}-1.93\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4990ms 0.4063ms 2.4614 KOps/s 2.4244 KOps/s $\color{#35bf28}+1.53\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 30.0605ms 29.4192ms 33.9914 Ops/s 42.7032 Ops/s $\textbf{\color{#d91a1a}-20.40\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.8286ms 1.6990ms 588.5845 Ops/s 587.2669 Ops/s $\color{#35bf28}+0.22\%$
test_dqn_speed[False-None] 1.4872ms 1.3684ms 730.7628 Ops/s 713.2190 Ops/s $\color{#35bf28}+2.46\%$
test_dqn_speed[False-backward] 1.9356ms 1.8641ms 536.4534 Ops/s 521.5353 Ops/s $\color{#35bf28}+2.86\%$
test_dqn_speed[True-None] 0.6757ms 0.5349ms 1.8694 KOps/s 1.8456 KOps/s $\color{#35bf28}+1.29\%$
test_dqn_speed[True-backward] 1.1135ms 0.9686ms 1.0325 KOps/s 1.0007 KOps/s $\color{#35bf28}+3.17\%$
test_dqn_speed[reduce-overhead-None] 0.5678ms 0.5161ms 1.9376 KOps/s 1.8410 KOps/s $\textbf{\color{#35bf28}+5.25\%}$
test_dqn_speed[reduce-overhead-backward] 0.9908ms 0.9554ms 1.0466 KOps/s 1.0443 KOps/s $\color{#35bf28}+0.23\%$
test_ddpg_speed[False-None] 3.5322ms 2.8409ms 352.0007 Ops/s 352.6440 Ops/s $\color{#d91a1a}-0.18\%$
test_ddpg_speed[False-backward] 4.1428ms 4.0280ms 248.2635 Ops/s 248.8695 Ops/s $\color{#d91a1a}-0.24\%$
test_ddpg_speed[True-None] 1.6076ms 1.3688ms 730.5741 Ops/s 719.5485 Ops/s $\color{#35bf28}+1.53\%$
test_ddpg_speed[True-backward] 2.5070ms 2.3687ms 422.1778 Ops/s 415.8319 Ops/s $\color{#35bf28}+1.53\%$
test_ddpg_speed[reduce-overhead-None] 2.4216ms 1.3946ms 717.0283 Ops/s 723.5195 Ops/s $\color{#d91a1a}-0.90\%$
test_ddpg_speed[reduce-overhead-backward] 2.4289ms 2.3322ms 428.7846 Ops/s 417.7892 Ops/s $\color{#35bf28}+2.63\%$
test_sac_speed[False-None] 9.1961ms 7.8084ms 128.0679 Ops/s 126.3661 Ops/s $\color{#35bf28}+1.35\%$
test_sac_speed[False-backward] 11.3097ms 10.9107ms 91.6530 Ops/s 89.7404 Ops/s $\color{#35bf28}+2.13\%$
test_sac_speed[True-None] 2.3064ms 2.1017ms 475.8047 Ops/s 458.4890 Ops/s $\color{#35bf28}+3.78\%$
test_sac_speed[True-backward] 4.0256ms 3.9171ms 255.2935 Ops/s 233.3012 Ops/s $\textbf{\color{#35bf28}+9.43\%}$
test_sac_speed[reduce-overhead-None] 2.3567ms 2.1289ms 469.7347 Ops/s 442.9436 Ops/s $\textbf{\color{#35bf28}+6.05\%}$
test_sac_speed[reduce-overhead-backward] 4.7882ms 4.4367ms 225.3952 Ops/s 251.0764 Ops/s $\textbf{\color{#d91a1a}-10.23\%}$
test_redq_speed[False-None] 12.5689ms 10.4347ms 95.8342 Ops/s 98.0470 Ops/s $\color{#d91a1a}-2.26\%$
test_redq_speed[False-backward] 18.3640ms 17.8763ms 55.9400 Ops/s 56.0088 Ops/s $\color{#d91a1a}-0.12\%$
test_redq_speed[True-None] 4.8645ms 4.5732ms 218.6676 Ops/s 217.7499 Ops/s $\color{#35bf28}+0.42\%$
test_redq_speed[True-backward] 10.1539ms 9.8432ms 101.5932 Ops/s 106.6269 Ops/s $\color{#d91a1a}-4.72\%$
test_redq_speed[reduce-overhead-None] 4.6725ms 4.3978ms 227.3864 Ops/s 229.7539 Ops/s $\color{#d91a1a}-1.03\%$
test_redq_speed[reduce-overhead-backward] 12.1914ms 9.9596ms 100.4058 Ops/s 97.1744 Ops/s $\color{#35bf28}+3.33\%$
test_redq_deprec_speed[False-None] 13.6369ms 11.0590ms 90.4243 Ops/s 93.3085 Ops/s $\color{#d91a1a}-3.09\%$
test_redq_deprec_speed[False-backward] 16.3813ms 15.9235ms 62.8004 Ops/s 65.3906 Ops/s $\color{#d91a1a}-3.96\%$
test_redq_deprec_speed[True-None] 3.8524ms 3.6311ms 275.4009 Ops/s 268.8265 Ops/s $\color{#35bf28}+2.45\%$
test_redq_deprec_speed[True-backward] 7.9731ms 7.6606ms 130.5380 Ops/s 130.4093 Ops/s $\color{#35bf28}+0.10\%$
test_redq_deprec_speed[reduce-overhead-None] 3.8103ms 3.6510ms 273.8998 Ops/s 280.9349 Ops/s $\color{#d91a1a}-2.50\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.0891ms 7.7642ms 128.7967 Ops/s 130.6407 Ops/s $\color{#d91a1a}-1.41\%$
test_td3_speed[False-None] 8.3129ms 7.9328ms 126.0584 Ops/s 125.1011 Ops/s $\color{#35bf28}+0.77\%$
test_td3_speed[False-backward] 11.1054ms 10.6366ms 94.0151 Ops/s 93.6276 Ops/s $\color{#35bf28}+0.41\%$
test_td3_speed[True-None] 1.8596ms 1.8139ms 551.2961 Ops/s 554.8041 Ops/s $\color{#d91a1a}-0.63\%$
test_td3_speed[True-backward] 3.7811ms 3.5825ms 279.1345 Ops/s 239.2835 Ops/s $\textbf{\color{#35bf28}+16.65\%}$
test_td3_speed[reduce-overhead-None] 1.8281ms 1.7731ms 563.9866 Ops/s 552.7204 Ops/s $\color{#35bf28}+2.04\%$
test_td3_speed[reduce-overhead-backward] 4.2490ms 3.6595ms 273.2616 Ops/s 276.4670 Ops/s $\color{#d91a1a}-1.16\%$
test_cql_speed[False-None] 27.9982ms 25.7537ms 38.8293 Ops/s 38.5693 Ops/s $\color{#35bf28}+0.67\%$
test_cql_speed[False-backward] 35.5265ms 34.8446ms 28.6989 Ops/s 28.5345 Ops/s $\color{#35bf28}+0.58\%$
test_cql_speed[True-None] 13.6614ms 12.5508ms 79.6764 Ops/s 81.4020 Ops/s $\color{#d91a1a}-2.12\%$
test_cql_speed[True-backward] 19.1152ms 18.6478ms 53.6257 Ops/s 56.4730 Ops/s $\textbf{\color{#d91a1a}-5.04\%}$
test_cql_speed[reduce-overhead-None] 12.9262ms 12.5732ms 79.5344 Ops/s 81.3636 Ops/s $\color{#d91a1a}-2.25\%$
test_cql_speed[reduce-overhead-backward] 18.6754ms 18.2232ms 54.8752 Ops/s 56.7777 Ops/s $\color{#d91a1a}-3.35\%$
test_a2c_speed[False-None] 5.5222ms 5.3359ms 187.4110 Ops/s 185.0679 Ops/s $\color{#35bf28}+1.27\%$
test_a2c_speed[False-backward] 11.9385ms 11.6622ms 85.7468 Ops/s 82.5892 Ops/s $\color{#35bf28}+3.82\%$
test_a2c_speed[True-None] 3.8074ms 3.6749ms 272.1155 Ops/s 287.0258 Ops/s $\textbf{\color{#d91a1a}-5.19\%}$
test_a2c_speed[True-backward] 8.8194ms 8.5841ms 116.4949 Ops/s 118.7776 Ops/s $\color{#d91a1a}-1.92\%$
test_a2c_speed[reduce-overhead-None] 3.8049ms 3.6593ms 273.2785 Ops/s 285.6424 Ops/s $\color{#d91a1a}-4.33\%$
test_a2c_speed[reduce-overhead-backward] 8.9799ms 8.7270ms 114.5867 Ops/s 117.6663 Ops/s $\color{#d91a1a}-2.62\%$
test_ppo_speed[False-None] 6.0047ms 5.7533ms 173.8133 Ops/s 178.3804 Ops/s $\color{#d91a1a}-2.56\%$
test_ppo_speed[False-backward] 12.4783ms 12.1176ms 82.5246 Ops/s 81.2077 Ops/s $\color{#35bf28}+1.62\%$
test_ppo_speed[True-None] 3.9326ms 3.5732ms 279.8618 Ops/s 298.5032 Ops/s $\textbf{\color{#d91a1a}-6.24\%}$
test_ppo_speed[True-backward] 8.8680ms 8.4627ms 118.1657 Ops/s 120.6404 Ops/s $\color{#d91a1a}-2.05\%$
test_ppo_speed[reduce-overhead-None] 3.7796ms 3.5838ms 279.0299 Ops/s 298.4607 Ops/s $\textbf{\color{#d91a1a}-6.51\%}$
test_ppo_speed[reduce-overhead-backward] 8.9037ms 8.6517ms 115.5837 Ops/s 110.8614 Ops/s $\color{#35bf28}+4.26\%$
test_reinforce_speed[False-None] 5.9287ms 4.4844ms 222.9952 Ops/s 217.8833 Ops/s $\color{#35bf28}+2.35\%$
test_reinforce_speed[False-backward] 7.6696ms 7.3448ms 136.1503 Ops/s 135.9687 Ops/s $\color{#35bf28}+0.13\%$
test_reinforce_speed[True-None] 3.3024ms 2.8812ms 347.0778 Ops/s 371.6247 Ops/s $\textbf{\color{#d91a1a}-6.61\%}$
test_reinforce_speed[True-backward] 8.2937ms 7.8120ms 128.0082 Ops/s 134.6422 Ops/s $\color{#d91a1a}-4.93\%$
test_reinforce_speed[reduce-overhead-None] 3.1669ms 2.8741ms 347.9398 Ops/s 338.7640 Ops/s $\color{#35bf28}+2.71\%$
test_reinforce_speed[reduce-overhead-backward] 8.0414ms 7.8395ms 127.5598 Ops/s 126.5300 Ops/s $\color{#35bf28}+0.81\%$
test_iql_speed[False-None] 24.8397ms 20.0192ms 49.9521 Ops/s 51.0334 Ops/s $\color{#d91a1a}-2.12\%$
test_iql_speed[False-backward] 30.9294ms 30.2350ms 33.0743 Ops/s 33.7582 Ops/s $\color{#d91a1a}-2.03\%$
test_iql_speed[True-None] 9.0820ms 8.6408ms 115.7304 Ops/s 125.1073 Ops/s $\textbf{\color{#d91a1a}-7.50\%}$
test_iql_speed[True-backward] 17.1626ms 16.8287ms 59.4223 Ops/s 59.9140 Ops/s $\color{#d91a1a}-0.82\%$
test_iql_speed[reduce-overhead-None] 9.0831ms 8.7032ms 114.9005 Ops/s 115.3699 Ops/s $\color{#d91a1a}-0.41\%$
test_iql_speed[reduce-overhead-backward] 17.5982ms 17.2315ms 58.0332 Ops/s 55.3969 Ops/s $\color{#35bf28}+4.76\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1896ms 6.0399ms 165.5662 Ops/s 165.3808 Ops/s $\color{#35bf28}+0.11\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8002ms 0.2844ms 3.5159 KOps/s 2.9046 KOps/s $\textbf{\color{#35bf28}+21.04\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5626ms 0.2912ms 3.4338 KOps/s 2.9041 KOps/s $\textbf{\color{#35bf28}+18.24\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2544ms 5.9602ms 167.7791 Ops/s 174.4205 Ops/s $\color{#d91a1a}-3.81\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1489ms 0.3671ms 2.7240 KOps/s 2.9113 KOps/s $\textbf{\color{#d91a1a}-6.43\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6845ms 0.3537ms 2.8272 KOps/s 3.6174 KOps/s $\textbf{\color{#d91a1a}-21.84\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7709ms 1.4027ms 712.9127 Ops/s 693.8200 Ops/s $\color{#35bf28}+2.75\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5801ms 1.3259ms 754.1952 Ops/s 752.9374 Ops/s $\color{#35bf28}+0.17\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3368ms 6.0982ms 163.9822 Ops/s 167.9887 Ops/s $\color{#d91a1a}-2.39\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8597ms 0.5199ms 1.9235 KOps/s 2.0430 KOps/s $\textbf{\color{#d91a1a}-5.85\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7793ms 0.4992ms 2.0032 KOps/s 2.1115 KOps/s $\textbf{\color{#d91a1a}-5.13\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9621ms 5.8685ms 170.4001 Ops/s 171.2308 Ops/s $\color{#d91a1a}-0.49\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8335ms 0.3689ms 2.7107 KOps/s 2.7859 KOps/s $\color{#d91a1a}-2.70\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5598ms 0.3505ms 2.8534 KOps/s 2.9143 KOps/s $\color{#d91a1a}-2.09\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0574ms 5.8307ms 171.5069 Ops/s 172.5239 Ops/s $\color{#d91a1a}-0.59\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.3066ms 0.3579ms 2.7938 KOps/s 2.8514 KOps/s $\color{#d91a1a}-2.02\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5607ms 0.3395ms 2.9454 KOps/s 2.9803 KOps/s $\color{#d91a1a}-1.17\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2698ms 6.0944ms 164.0843 Ops/s 167.3909 Ops/s $\color{#d91a1a}-1.98\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.0247ms 0.4899ms 2.0411 KOps/s 2.1643 KOps/s $\textbf{\color{#d91a1a}-5.69\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8404ms 0.5108ms 1.9579 KOps/s 2.3531 KOps/s $\textbf{\color{#d91a1a}-16.80\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.7074s 19.3178ms 51.7658 Ops/s 194.8870 Ops/s $\textbf{\color{#d91a1a}-73.44\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.1570ms 2.0934ms 477.6834 Ops/s 425.7392 Ops/s $\textbf{\color{#35bf28}+12.20\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.0120ms 1.1781ms 848.8358 Ops/s 915.9201 Ops/s $\textbf{\color{#d91a1a}-7.32\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.7595ms 5.1713ms 193.3764 Ops/s 49.2018 Ops/s $\textbf{\color{#35bf28}+293.03\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 10.1553ms 2.0799ms 480.7960 Ops/s 681.8365 Ops/s $\textbf{\color{#d91a1a}-29.49\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 3.4598ms 1.1180ms 894.4338 Ops/s 769.4510 Ops/s $\textbf{\color{#35bf28}+16.24\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 9.5326ms 5.2998ms 188.6858 Ops/s 187.8406 Ops/s $\color{#35bf28}+0.45\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.9618ms 2.1912ms 456.3666 Ops/s 449.0726 Ops/s $\color{#35bf28}+1.62\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 9.4314ms 1.4063ms 711.0942 Ops/s 719.5886 Ops/s $\color{#d91a1a}-1.18\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 35.5657ms 33.0333ms 30.2725 Ops/s 29.2532 Ops/s $\color{#35bf28}+3.48\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.1739ms 17.4080ms 57.4447 Ops/s 55.5903 Ops/s $\color{#35bf28}+3.34\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 36.9358ms 34.1967ms 29.2426 Ops/s 26.1472 Ops/s $\textbf{\color{#35bf28}+11.84\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.6389ms 17.2973ms 57.8124 Ops/s 54.0092 Ops/s $\textbf{\color{#35bf28}+7.04\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 41.1719ms 36.6448ms 27.2890 Ops/s 26.4737 Ops/s $\color{#35bf28}+3.08\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 23.1771ms 19.4104ms 51.5188 Ops/s 49.8697 Ops/s $\color{#35bf28}+3.31\%$

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: 1053bd7
Pull-Request: #3264
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: 1053bd7
Pull-Request: #3264
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: 6ea3fd7
Pull-Request: #3264
[ghstack-poisoned]
@vmoens vmoens added the CI Has to do with CI setup (e.g. wheels & builds, tests...) label Dec 18, 2025
vmoens added a commit that referenced this pull request Dec 18, 2025
ghstack-source-id: e792387
Pull-Request: #3264
@vmoens vmoens merged commit 00a7b9d into gh/vmoens/173/base Dec 18, 2025
98 of 106 checks passed
@vmoens vmoens deleted the gh/vmoens/173/head branch December 18, 2025 16:09
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