Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Dec 31, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Dec 31, 2025
ghstack-source-id: ba49af9
Pull-Request: #3280
@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/3280

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:

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 Dec 31, 2025
vmoens added a commit that referenced this pull request Dec 31, 2025
ghstack-source-id: ba49af9
Pull-Request: #3280
@vmoens vmoens merged commit 9f8a805 into gh/vmoens/168/base Dec 31, 2025
52 of 74 checks passed
@vmoens vmoens deleted the gh/vmoens/168/head branch December 31, 2025 09:21
@github-actions
Copy link

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

Total Benchmarks: 164. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 82.1842μs 81.3669μs 12.2900 KOps/s 12.4236 KOps/s $\color{#d91a1a}-1.07\%$
test_tensor_to_bytestream_speed[torch.save] 0.1408ms 0.1405ms 7.1161 KOps/s 7.0902 KOps/s $\color{#35bf28}+0.37\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1192s 0.1191s 8.3958 Ops/s 8.2653 Ops/s $\color{#35bf28}+1.58\%$
test_tensor_to_bytestream_speed[numpy] 2.6747μs 2.6660μs 375.0884 KOps/s 375.0574 KOps/s $+0.01\%$
test_tensor_to_bytestream_speed[safetensors] 38.4347μs 38.0720μs 26.2660 KOps/s 25.6041 KOps/s $\color{#35bf28}+2.59\%$
test_simple 0.5650s 0.5617s 1.7803 Ops/s 1.7211 Ops/s $\color{#35bf28}+3.44\%$
test_transformed 1.1411s 1.1368s 0.8796 Ops/s 0.8596 Ops/s $\color{#35bf28}+2.33\%$
test_serial 1.7004s 1.6850s 0.5935 Ops/s 0.5870 Ops/s $\color{#35bf28}+1.10\%$
test_parallel 1.4137s 1.2100s 0.8265 Ops/s 0.8443 Ops/s $\color{#d91a1a}-2.11\%$
test_step_mdp_speed[True-True-True-True-True] 0.2007ms 44.0785μs 22.6868 KOps/s 22.6903 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[True-True-True-True-False] 62.6520μs 24.9991μs 40.0014 KOps/s 40.1299 KOps/s $\color{#d91a1a}-0.32\%$
test_step_mdp_speed[True-True-True-False-True] 55.7610μs 25.3513μs 39.4458 KOps/s 40.1421 KOps/s $\color{#d91a1a}-1.73\%$
test_step_mdp_speed[True-True-True-False-False] 40.2210μs 14.0427μs 71.2112 KOps/s 73.1878 KOps/s $\color{#d91a1a}-2.70\%$
test_step_mdp_speed[True-True-False-True-True] 93.2010μs 48.5940μs 20.5787 KOps/s 20.8253 KOps/s $\color{#d91a1a}-1.18\%$
test_step_mdp_speed[True-True-False-True-False] 64.4410μs 27.9808μs 35.7388 KOps/s 35.4516 KOps/s $\color{#35bf28}+0.81\%$
test_step_mdp_speed[True-True-False-False-True] 86.2210μs 28.2685μs 35.3751 KOps/s 35.2479 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[True-True-False-False-False] 51.6410μs 16.7704μs 59.6289 KOps/s 60.6024 KOps/s $\color{#d91a1a}-1.61\%$
test_step_mdp_speed[True-False-True-True-True] 94.0420μs 50.7089μs 19.7204 KOps/s 19.4658 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[True-False-True-True-False] 76.7520μs 30.7457μs 32.5249 KOps/s 32.6987 KOps/s $\color{#d91a1a}-0.53\%$
test_step_mdp_speed[True-False-True-False-True] 69.6110μs 27.7526μs 36.0327 KOps/s 36.4733 KOps/s $\color{#d91a1a}-1.21\%$
test_step_mdp_speed[True-False-True-False-False] 45.3410μs 16.6075μs 60.2137 KOps/s 60.7551 KOps/s $\color{#d91a1a}-0.89\%$
test_step_mdp_speed[True-False-False-True-True] 97.3120μs 52.6224μs 19.0033 KOps/s 19.0604 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-False-False-True-False] 68.3610μs 33.3956μs 29.9441 KOps/s 29.7697 KOps/s $\color{#35bf28}+0.59\%$
test_step_mdp_speed[True-False-False-False-True] 67.7410μs 30.7703μs 32.4988 KOps/s 33.0741 KOps/s $\color{#d91a1a}-1.74\%$
test_step_mdp_speed[True-False-False-False-False] 49.1210μs 19.3170μs 51.7680 KOps/s 51.7595 KOps/s $\color{#35bf28}+0.02\%$
test_step_mdp_speed[False-True-True-True-True] 83.0910μs 49.6191μs 20.1535 KOps/s 19.6577 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[False-True-True-True-False] 73.1610μs 30.8779μs 32.3856 KOps/s 32.3358 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[False-True-True-False-True] 2.2803ms 31.5781μs 31.6675 KOps/s 31.1884 KOps/s $\color{#35bf28}+1.54\%$
test_step_mdp_speed[False-True-True-False-False] 49.0310μs 18.4915μs 54.0789 KOps/s 54.0444 KOps/s $\color{#35bf28}+0.06\%$
test_step_mdp_speed[False-True-False-True-True] 0.1125ms 53.5633μs 18.6695 KOps/s 18.5619 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[False-True-False-True-False] 64.5710μs 33.3865μs 29.9522 KOps/s 29.8066 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[False-True-False-False-True] 64.8120μs 33.9821μs 29.4272 KOps/s 29.1284 KOps/s $\color{#35bf28}+1.03\%$
test_step_mdp_speed[False-True-False-False-False] 63.7720μs 21.0171μs 47.5802 KOps/s 47.4783 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[False-False-True-True-True] 90.0010μs 56.1172μs 17.8198 KOps/s 17.5950 KOps/s $\color{#35bf28}+1.28\%$
test_step_mdp_speed[False-False-True-True-False] 70.5010μs 36.3334μs 27.5229 KOps/s 27.6055 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[False-False-True-False-True] 78.1710μs 34.2898μs 29.1632 KOps/s 29.0883 KOps/s $\color{#35bf28}+0.26\%$
test_step_mdp_speed[False-False-True-False-False] 67.0610μs 20.8473μs 47.9678 KOps/s 47.1996 KOps/s $\color{#35bf28}+1.63\%$
test_step_mdp_speed[False-False-False-True-True] 0.1002ms 57.4969μs 17.3922 KOps/s 17.0523 KOps/s $\color{#35bf28}+1.99\%$
test_step_mdp_speed[False-False-False-True-False] 77.6320μs 38.4943μs 25.9779 KOps/s 25.7271 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[False-False-False-False-True] 76.6210μs 35.5564μs 28.1243 KOps/s 27.7886 KOps/s $\color{#35bf28}+1.21\%$
test_step_mdp_speed[False-False-False-False-False] 56.2210μs 23.2824μs 42.9509 KOps/s 42.7123 KOps/s $\color{#35bf28}+0.56\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8913s 0.7928s 1.2613 Ops/s 1.2690 Ops/s $\color{#d91a1a}-0.61\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7479s 0.6503s 1.5377 Ops/s 1.5324 Ops/s $\color{#35bf28}+0.34\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.8108s 1.7311s 0.5777 Ops/s 0.5799 Ops/s $\color{#d91a1a}-0.39\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5721s 1.4977s 0.6677 Ops/s 0.6716 Ops/s $\color{#d91a1a}-0.58\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0832s 1.9943s 0.5014 Ops/s 0.5081 Ops/s $\color{#d91a1a}-1.30\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.8388s 1.7627s 0.5673 Ops/s 0.5748 Ops/s $\color{#d91a1a}-1.30\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8416s 4.7613s 0.2100 Ops/s 0.2092 Ops/s $\color{#35bf28}+0.38\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.6370s 4.5335s 0.2206 Ops/s 0.2227 Ops/s $\color{#d91a1a}-0.96\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 2.0954s 1.9993s 0.5002 Ops/s 0.5039 Ops/s $\color{#d91a1a}-0.74\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.8477s 1.7066s 0.5860 Ops/s 0.5803 Ops/s $\color{#35bf28}+0.98\%$
test_values[generalized_advantage_estimate-True-True] 10.6971ms 10.2439ms 97.6192 Ops/s 97.3763 Ops/s $\color{#35bf28}+0.25\%$
test_values[vec_generalized_advantage_estimate-True-True] 19.5498ms 17.8129ms 56.1391 Ops/s 55.7834 Ops/s $\color{#35bf28}+0.64\%$
test_values[td0_return_estimate-False-False] 0.2297ms 0.1319ms 7.5839 KOps/s 7.9029 KOps/s $\color{#d91a1a}-4.04\%$
test_values[td1_return_estimate-False-False] 28.1008ms 27.7198ms 36.0753 Ops/s 35.5096 Ops/s $\color{#35bf28}+1.59\%$
test_values[vec_td1_return_estimate-False-False] 18.9645ms 17.8851ms 55.9124 Ops/s 55.8265 Ops/s $\color{#35bf28}+0.15\%$
test_values[td_lambda_return_estimate-True-False] 41.4744ms 40.9528ms 24.4184 Ops/s 24.1241 Ops/s $\color{#35bf28}+1.22\%$
test_values[vec_td_lambda_return_estimate-True-False] 18.8062ms 17.6891ms 56.5319 Ops/s 55.7207 Ops/s $\color{#35bf28}+1.46\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.2097ms 9.1178ms 109.6754 Ops/s 107.9968 Ops/s $\color{#35bf28}+1.55\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.8835ms 1.4989ms 667.1650 Ops/s 657.3906 Ops/s $\color{#35bf28}+1.49\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4713ms 0.4186ms 2.3891 KOps/s 2.3766 KOps/s $\color{#35bf28}+0.52\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 35.1543ms 34.3829ms 29.0843 Ops/s 28.6521 Ops/s $\color{#35bf28}+1.51\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 2.3514ms 1.8271ms 547.3187 Ops/s 552.1027 Ops/s $\color{#d91a1a}-0.87\%$
test_dqn_speed[False-None] 1.8542ms 1.4184ms 705.0295 Ops/s 701.0104 Ops/s $\color{#35bf28}+0.57\%$
test_dqn_speed[False-backward] 1.9969ms 1.9295ms 518.2713 Ops/s 520.6362 Ops/s $\color{#d91a1a}-0.45\%$
test_dqn_speed[True-None] 0.9931ms 0.5351ms 1.8688 KOps/s 1.8641 KOps/s $\color{#35bf28}+0.25\%$
test_dqn_speed[True-backward] 1.0464ms 0.9874ms 1.0127 KOps/s 989.9518 Ops/s $\color{#35bf28}+2.30\%$
test_dqn_speed[reduce-overhead-None] 0.6533ms 0.5368ms 1.8628 KOps/s 1.8489 KOps/s $\color{#35bf28}+0.76\%$
test_dqn_speed[reduce-overhead-backward] 1.0992ms 1.0058ms 994.2403 Ops/s 894.6989 Ops/s $\textbf{\color{#35bf28}+11.13\%}$
test_ddpg_speed[False-None] 3.2819ms 2.9641ms 337.3668 Ops/s 340.6230 Ops/s $\color{#d91a1a}-0.96\%$
test_ddpg_speed[False-backward] 4.2660ms 4.1668ms 239.9897 Ops/s 239.1183 Ops/s $\color{#35bf28}+0.36\%$
test_ddpg_speed[True-None] 1.5264ms 1.4043ms 712.0888 Ops/s 708.1973 Ops/s $\color{#35bf28}+0.55\%$
test_ddpg_speed[True-backward] 2.5796ms 2.4189ms 413.4041 Ops/s 417.4034 Ops/s $\color{#d91a1a}-0.96\%$
test_ddpg_speed[reduce-overhead-None] 1.6121ms 1.4029ms 712.8297 Ops/s 714.3767 Ops/s $\color{#d91a1a}-0.22\%$
test_ddpg_speed[reduce-overhead-backward] 2.4787ms 2.4018ms 416.3541 Ops/s 415.3761 Ops/s $\color{#35bf28}+0.24\%$
test_sac_speed[False-None] 8.6265ms 8.2168ms 121.7025 Ops/s 123.9410 Ops/s $\color{#d91a1a}-1.81\%$
test_sac_speed[False-backward] 12.3673ms 11.5372ms 86.6764 Ops/s 88.2447 Ops/s $\color{#d91a1a}-1.78\%$
test_sac_speed[True-None] 2.4071ms 2.1734ms 460.1045 Ops/s 464.9135 Ops/s $\color{#d91a1a}-1.03\%$
test_sac_speed[True-backward] 4.3037ms 4.0879ms 244.6261 Ops/s 208.9343 Ops/s $\textbf{\color{#35bf28}+17.08\%}$
test_sac_speed[reduce-overhead-None] 2.4970ms 2.2120ms 452.0875 Ops/s 450.1754 Ops/s $\color{#35bf28}+0.42\%$
test_sac_speed[reduce-overhead-backward] 4.2459ms 4.1009ms 243.8508 Ops/s 243.4200 Ops/s $\color{#35bf28}+0.18\%$
test_redq_speed[False-None] 11.0777ms 10.4466ms 95.7248 Ops/s 95.3869 Ops/s $\color{#35bf28}+0.35\%$
test_redq_speed[False-backward] 23.3527ms 18.2331ms 54.8452 Ops/s 54.5876 Ops/s $\color{#35bf28}+0.47\%$
test_redq_speed[True-None] 4.7414ms 4.4626ms 224.0833 Ops/s 231.1712 Ops/s $\color{#d91a1a}-3.07\%$
test_redq_speed[True-backward] 10.1612ms 9.8509ms 101.5139 Ops/s 97.9261 Ops/s $\color{#35bf28}+3.66\%$
test_redq_speed[reduce-overhead-None] 4.5413ms 4.3693ms 228.8708 Ops/s 226.0713 Ops/s $\color{#35bf28}+1.24\%$
test_redq_speed[reduce-overhead-backward] 10.3066ms 10.0413ms 99.5886 Ops/s 101.9401 Ops/s $\color{#d91a1a}-2.31\%$
test_redq_deprec_speed[False-None] 11.8316ms 11.3196ms 88.3423 Ops/s 89.7242 Ops/s $\color{#d91a1a}-1.54\%$
test_redq_deprec_speed[False-backward] 17.3183ms 16.3781ms 61.0570 Ops/s 63.0965 Ops/s $\color{#d91a1a}-3.23\%$
test_redq_deprec_speed[True-None] 4.1317ms 3.6968ms 270.5066 Ops/s 271.4786 Ops/s $\color{#d91a1a}-0.36\%$
test_redq_deprec_speed[True-backward] 7.9728ms 7.7159ms 129.6023 Ops/s 112.3976 Ops/s $\textbf{\color{#35bf28}+15.31\%}$
test_redq_deprec_speed[reduce-overhead-None] 3.8509ms 3.6357ms 275.0474 Ops/s 259.6158 Ops/s $\textbf{\color{#35bf28}+5.94\%}$
test_redq_deprec_speed[reduce-overhead-backward] 7.9329ms 7.6841ms 130.1382 Ops/s 129.6388 Ops/s $\color{#35bf28}+0.39\%$
test_td3_speed[False-None] 8.4066ms 8.2262ms 121.5630 Ops/s 123.4201 Ops/s $\color{#d91a1a}-1.50\%$
test_td3_speed[False-backward] 11.8396ms 11.2014ms 89.2743 Ops/s 91.0378 Ops/s $\color{#d91a1a}-1.94\%$
test_td3_speed[True-None] 1.8868ms 1.8385ms 543.9127 Ops/s 526.8301 Ops/s $\color{#35bf28}+3.24\%$
test_td3_speed[True-backward] 3.7461ms 3.6233ms 275.9893 Ops/s 261.5857 Ops/s $\textbf{\color{#35bf28}+5.51\%}$
test_td3_speed[reduce-overhead-None] 1.8774ms 1.8224ms 548.7367 Ops/s 532.2199 Ops/s $\color{#35bf28}+3.10\%$
test_td3_speed[reduce-overhead-backward] 3.8268ms 3.7191ms 268.8796 Ops/s 237.1946 Ops/s $\textbf{\color{#35bf28}+13.36\%}$
test_cql_speed[False-None] 27.0698ms 26.1828ms 38.1930 Ops/s 37.9825 Ops/s $\color{#35bf28}+0.55\%$
test_cql_speed[False-backward] 40.3000ms 36.2432ms 27.5914 Ops/s 27.7049 Ops/s $\color{#d91a1a}-0.41\%$
test_cql_speed[True-None] 15.1034ms 12.4164ms 80.5389 Ops/s 78.2498 Ops/s $\color{#35bf28}+2.93\%$
test_cql_speed[True-backward] 18.9495ms 18.3971ms 54.3564 Ops/s 54.2368 Ops/s $\color{#35bf28}+0.22\%$
test_cql_speed[reduce-overhead-None] 12.7760ms 12.5212ms 79.8646 Ops/s 77.4716 Ops/s $\color{#35bf28}+3.09\%$
test_cql_speed[reduce-overhead-backward] 19.0463ms 18.4765ms 54.1229 Ops/s 55.5673 Ops/s $\color{#d91a1a}-2.60\%$
test_a2c_speed[False-None] 5.8930ms 5.4679ms 182.8851 Ops/s 178.7464 Ops/s $\color{#35bf28}+2.32\%$
test_a2c_speed[False-backward] 12.2229ms 11.9364ms 83.7773 Ops/s 82.8125 Ops/s $\color{#35bf28}+1.17\%$
test_a2c_speed[True-None] 4.1256ms 3.6939ms 270.7164 Ops/s 263.4311 Ops/s $\color{#35bf28}+2.77\%$
test_a2c_speed[True-backward] 9.2179ms 8.5932ms 116.3708 Ops/s 110.5940 Ops/s $\textbf{\color{#35bf28}+5.22\%}$
test_a2c_speed[reduce-overhead-None] 3.9886ms 3.7338ms 267.8203 Ops/s 268.7368 Ops/s $\color{#d91a1a}-0.34\%$
test_a2c_speed[reduce-overhead-backward] 9.1768ms 8.7277ms 114.5775 Ops/s 103.3461 Ops/s $\textbf{\color{#35bf28}+10.87\%}$
test_ppo_speed[False-None] 6.5900ms 5.9957ms 166.7865 Ops/s 166.7251 Ops/s $\color{#35bf28}+0.04\%$
test_ppo_speed[False-backward] 13.3396ms 12.7138ms 78.6546 Ops/s 78.8343 Ops/s $\color{#d91a1a}-0.23\%$
test_ppo_speed[True-None] 4.0490ms 3.6637ms 272.9516 Ops/s 271.0024 Ops/s $\color{#35bf28}+0.72\%$
test_ppo_speed[True-backward] 8.6906ms 8.4833ms 117.8792 Ops/s 115.9266 Ops/s $\color{#35bf28}+1.68\%$
test_ppo_speed[reduce-overhead-None] 3.7894ms 3.6139ms 276.7094 Ops/s 274.7266 Ops/s $\color{#35bf28}+0.72\%$
test_ppo_speed[reduce-overhead-backward] 9.1790ms 8.7282ms 114.5707 Ops/s 107.8180 Ops/s $\textbf{\color{#35bf28}+6.26\%}$
test_reinforce_speed[False-None] 5.0586ms 4.6062ms 217.0994 Ops/s 216.6021 Ops/s $\color{#35bf28}+0.23\%$
test_reinforce_speed[False-backward] 7.6443ms 7.4447ms 134.3231 Ops/s 133.2834 Ops/s $\color{#35bf28}+0.78\%$
test_reinforce_speed[True-None] 3.4271ms 2.9289ms 341.4215 Ops/s 336.2593 Ops/s $\color{#35bf28}+1.54\%$
test_reinforce_speed[True-backward] 8.0415ms 7.7140ms 129.6338 Ops/s 102.9245 Ops/s $\textbf{\color{#35bf28}+25.95\%}$
test_reinforce_speed[reduce-overhead-None] 3.3652ms 2.9088ms 343.7805 Ops/s 342.6858 Ops/s $\color{#35bf28}+0.32\%$
test_reinforce_speed[reduce-overhead-backward] 8.2305ms 7.9627ms 125.5855 Ops/s 122.2127 Ops/s $\color{#35bf28}+2.76\%$
test_iql_speed[False-None] 26.4753ms 21.1252ms 47.3368 Ops/s 49.1058 Ops/s $\color{#d91a1a}-3.60\%$
test_iql_speed[False-backward] 31.7062ms 30.9234ms 32.3379 Ops/s 32.5502 Ops/s $\color{#d91a1a}-0.65\%$
test_iql_speed[True-None] 9.1013ms 8.5357ms 117.1549 Ops/s 116.4186 Ops/s $\color{#35bf28}+0.63\%$
test_iql_speed[True-backward] 17.1689ms 16.8244ms 59.4375 Ops/s 58.4911 Ops/s $\color{#35bf28}+1.62\%$
test_iql_speed[reduce-overhead-None] 8.8611ms 8.5792ms 116.5609 Ops/s 115.5681 Ops/s $\color{#35bf28}+0.86\%$
test_iql_speed[reduce-overhead-backward] 17.8292ms 17.3105ms 57.7685 Ops/s 58.7586 Ops/s $\color{#d91a1a}-1.69\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.0952ms 6.1093ms 163.6852 Ops/s 164.9062 Ops/s $\color{#d91a1a}-0.74\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6497ms 0.3625ms 2.7586 KOps/s 3.4245 KOps/s $\textbf{\color{#d91a1a}-19.45\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5776ms 0.2964ms 3.3737 KOps/s 3.7219 KOps/s $\textbf{\color{#d91a1a}-9.36\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0886ms 5.8197ms 171.8296 Ops/s 172.0340 Ops/s $\color{#d91a1a}-0.12\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7821ms 0.3571ms 2.8004 KOps/s 3.5063 KOps/s $\textbf{\color{#d91a1a}-20.13\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5506ms 0.3388ms 2.9519 KOps/s 3.7881 KOps/s $\textbf{\color{#d91a1a}-22.07\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6265ms 1.4270ms 700.7695 Ops/s 781.3842 Ops/s $\textbf{\color{#d91a1a}-10.32\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6000ms 1.3538ms 738.6554 Ops/s 828.4789 Ops/s $\textbf{\color{#d91a1a}-10.84\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.0319ms 6.1265ms 163.2266 Ops/s 165.9895 Ops/s $\color{#d91a1a}-1.66\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.8878ms 0.5083ms 1.9672 KOps/s 2.2839 KOps/s $\textbf{\color{#d91a1a}-13.87\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7276ms 0.4689ms 2.1324 KOps/s 2.3735 KOps/s $\textbf{\color{#d91a1a}-10.16\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9275ms 5.8087ms 172.1552 Ops/s 169.5193 Ops/s $\color{#35bf28}+1.55\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.9798ms 0.3649ms 2.7407 KOps/s 3.4341 KOps/s $\textbf{\color{#d91a1a}-20.19\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5933ms 0.3431ms 2.9143 KOps/s 2.7732 KOps/s $\textbf{\color{#35bf28}+5.09\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.9834ms 5.7428ms 174.1310 Ops/s 170.9888 Ops/s $\color{#35bf28}+1.84\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0099ms 0.2872ms 3.4815 KOps/s 3.1554 KOps/s $\textbf{\color{#35bf28}+10.33\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7249ms 0.2694ms 3.7120 KOps/s 3.3616 KOps/s $\textbf{\color{#35bf28}+10.42\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.0937ms 5.9836ms 167.1234 Ops/s 165.2417 Ops/s $\color{#35bf28}+1.14\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0432ms 0.4427ms 2.2588 KOps/s 2.2121 KOps/s $\color{#35bf28}+2.11\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6917ms 0.4626ms 2.1616 KOps/s 2.2642 KOps/s $\color{#d91a1a}-4.53\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.4182ms 4.9665ms 201.3502 Ops/s 194.1248 Ops/s $\color{#35bf28}+3.72\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.4017ms 2.3539ms 424.8321 Ops/s 417.4168 Ops/s $\color{#35bf28}+1.78\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 6.4430ms 1.2270ms 815.0237 Ops/s 806.0901 Ops/s $\color{#35bf28}+1.11\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.5187ms 5.0450ms 198.2152 Ops/s 51.0593 Ops/s $\textbf{\color{#35bf28}+288.21\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 10.1667ms 2.4032ms 416.1048 Ops/s 505.3757 Ops/s $\textbf{\color{#d91a1a}-17.66\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 3.4448ms 1.1211ms 891.9899 Ops/s 820.0290 Ops/s $\textbf{\color{#35bf28}+8.78\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.6141s 17.4682ms 57.2469 Ops/s 188.0872 Ops/s $\textbf{\color{#d91a1a}-69.56\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.2474ms 2.2100ms 452.4960 Ops/s 431.1353 Ops/s $\color{#35bf28}+4.95\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 8.6691ms 1.3540ms 738.5735 Ops/s 760.1969 Ops/s $\color{#d91a1a}-2.84\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 36.2161ms 34.4218ms 29.0514 Ops/s 28.5137 Ops/s $\color{#35bf28}+1.89\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.3918ms 17.9972ms 55.5643 Ops/s 56.2324 Ops/s $\color{#d91a1a}-1.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 37.0359ms 35.0964ms 28.4929 Ops/s 27.5802 Ops/s $\color{#35bf28}+3.31\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.0957ms 18.3118ms 54.6097 Ops/s 55.1695 Ops/s $\color{#d91a1a}-1.01\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 38.9130ms 37.2900ms 26.8168 Ops/s 26.0473 Ops/s $\color{#35bf28}+2.95\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.2647ms 19.9487ms 50.1285 Ops/s 49.2023 Ops/s $\color{#35bf28}+1.88\%$

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.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants