Skip to content

[Example] Dreamer: add IsaacLab training script and config #3477

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
vmoens wants to merge 1 commit into from
Closed

[Example] Dreamer: add IsaacLab training script and config #3477

vmoens wants to merge 1 commit into from
+753 −0

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Feb 10, 2026 •
edited
Loading

Stack from ghstack (oldest at bottom):

Add a complete Dreamer v1 training script for IsaacLab environments
with pixel-based TiledCamera observations:

  • dreamer_isaac.py: standalone training script with AppLauncher init
    (must happen before torch import), multi-GPU support (sim/train/eval),
    mixed precision (bfloat16), synchronous collect-train loop, async
    eval via RayEvalWorker, NaN detection, and LR warmup
  • config_isaac.yaml: Hydra config for pixel-based Isaac training
    (256 envs, 64x64 images, RSSM compile, bfloat16 autocast)

Co-authored-by: Cursor [email protected]

@vmoens
Update
a34c772 
[ghstack-poisoned]
@pytorch-bot
Copy link

pytorch-bot bot commented Feb 10, 2026 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/3477

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

❌ 3 New Failures

As of commit a34c772 with merge base 0bc6d20 (image):

NEW FAILURES - The following jobs have failed:

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

@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

⚠️ PR Title Label Error

Unknown or invalid prefix [Example].

Current title: [Example] Dreamer: add IsaacLab training script and config

Supported Prefixes (case-sensitive)

Your PR title must start with exactly one of these prefixes:

Prefix Label Applied Example
[BugFix] BugFix [BugFix] Fix memory leak in collector
[Feature] Feature [Feature] Add new optimizer
[Doc] or [Docs] Documentation [Doc] Update installation guide
[Refactor] Refactoring [Refactor] Clean up module imports
[CI] CI [CI] Fix workflow permissions
[Test] or [Tests] Tests [Tests] Add unit tests for buffer
[Environment] or [Environments] Environments [Environments] Add Gymnasium support
[Data] Data [Data] Fix replay buffer sampling
[Performance] or [Perf] Performance [Performance] Optimize tensor ops
[BC-Breaking] bc breaking [BC-Breaking] Remove deprecated API
[Deprecation] Deprecation [Deprecation] Mark old function

Note: Common variations like singular/plural are supported (e.g., [Doc] or [Docs]).

This was referenced Feb 10, 2026
Merged
Merged
Merged
Merged
Closed
Closed
Closed
@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

⚠️ PR Title Label Error

Unknown or invalid prefix [Example].

Current title: [Example] Dreamer: add IsaacLab training script and config

Supported Prefixes (case-sensitive)

Your PR title must start with exactly one of these prefixes:

Prefix Label Applied Example
[BugFix] BugFix [BugFix] Fix memory leak in collector
[Feature] Feature [Feature] Add new optimizer
[Doc] or [Docs] Documentation [Doc] Update installation guide
[Refactor] Refactoring [Refactor] Clean up module imports
[CI] CI [CI] Fix workflow permissions
[Test] or [Tests] Tests [Tests] Add unit tests for buffer
[Environment] or [Environments] Environments [Environments] Add Gymnasium support
[Data] Data [Data] Fix replay buffer sampling
[Performance] or [Perf] Performance [Performance] Optimize tensor ops
[BC-Breaking] bc breaking [BC-Breaking] Remove deprecated API
[Deprecation] Deprecation [Deprecation] Mark old function

Note: Common variations like singular/plural are supported (e.g., [Doc] or [Docs]).

@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 Feb 10, 2026
@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

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

Total Benchmarks: 173. Improved: $\large\color{#35bf28}21$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 81.2327μs 80.1967μs 12.4693 KOps/s 12.0365 KOps/s $\color{#35bf28}+3.60\%$
test_tensor_to_bytestream_speed[torch.save] 0.1383ms 0.1378ms 7.2566 KOps/s 6.9444 KOps/s $\color{#35bf28}+4.50\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1086s 0.1079s 9.2705 Ops/s 9.5145 Ops/s $\color{#d91a1a}-2.57\%$
test_tensor_to_bytestream_speed[numpy] 2.4064μs 2.3970μs 417.1874 KOps/s 405.8527 KOps/s $\color{#35bf28}+2.79\%$
test_tensor_to_bytestream_speed[safetensors] 36.0019μs 35.6233μs 28.0715 KOps/s 27.4579 KOps/s $\color{#35bf28}+2.23\%$
test_simple 0.5456s 0.5439s 1.8386 Ops/s 1.7685 Ops/s $\color{#35bf28}+3.97\%$
test_transformed 1.1256s 1.1226s 0.8908 Ops/s 0.8729 Ops/s $\color{#35bf28}+2.04\%$
test_serial 1.6544s 1.6505s 0.6059 Ops/s 0.5919 Ops/s $\color{#35bf28}+2.36\%$
test_parallel 1.1380s 1.0410s 0.9606 Ops/s 0.9610 Ops/s $\color{#d91a1a}-0.03\%$
test_step_mdp_speed[True-True-True-True-True] 0.1873ms 44.5480μs 22.4477 KOps/s 21.9911 KOps/s $\color{#35bf28}+2.08\%$
test_step_mdp_speed[True-True-True-True-False] 53.4900μs 25.5812μs 39.0913 KOps/s 38.4332 KOps/s $\color{#35bf28}+1.71\%$
test_step_mdp_speed[True-True-True-False-True] 60.1900μs 25.2310μs 39.6337 KOps/s 38.8509 KOps/s $\color{#35bf28}+2.01\%$
test_step_mdp_speed[True-True-True-False-False] 48.0500μs 13.7938μs 72.4965 KOps/s 70.9632 KOps/s $\color{#35bf28}+2.16\%$
test_step_mdp_speed[True-True-False-True-True] 89.7110μs 47.5809μs 21.0168 KOps/s 20.5285 KOps/s $\color{#35bf28}+2.38\%$
test_step_mdp_speed[True-True-False-True-False] 61.9800μs 27.7195μs 36.0757 KOps/s 35.2759 KOps/s $\color{#35bf28}+2.27\%$
test_step_mdp_speed[True-True-False-False-True] 66.3110μs 28.1393μs 35.5375 KOps/s 35.0548 KOps/s $\color{#35bf28}+1.38\%$
test_step_mdp_speed[True-True-False-False-False] 49.9710μs 16.7835μs 59.5825 KOps/s 59.0708 KOps/s $\color{#35bf28}+0.87\%$
test_step_mdp_speed[True-False-True-True-True] 0.1227ms 51.1484μs 19.5509 KOps/s 19.3546 KOps/s $\color{#35bf28}+1.01\%$
test_step_mdp_speed[True-False-True-True-False] 0.1105ms 30.1697μs 33.1459 KOps/s 32.0982 KOps/s $\color{#35bf28}+3.26\%$
test_step_mdp_speed[True-False-True-False-True] 69.1310μs 27.9570μs 35.7692 KOps/s 35.0955 KOps/s $\color{#35bf28}+1.92\%$
test_step_mdp_speed[True-False-True-False-False] 46.8310μs 16.5288μs 60.5004 KOps/s 58.5480 KOps/s $\color{#35bf28}+3.33\%$
test_step_mdp_speed[True-False-False-True-True] 99.0610μs 53.3014μs 18.7612 KOps/s 18.2688 KOps/s $\color{#35bf28}+2.70\%$
test_step_mdp_speed[True-False-False-True-False] 82.7910μs 33.6535μs 29.7146 KOps/s 29.4731 KOps/s $\color{#35bf28}+0.82\%$
test_step_mdp_speed[True-False-False-False-True] 93.6310μs 29.8349μs 33.5178 KOps/s 32.6162 KOps/s $\color{#35bf28}+2.76\%$
test_step_mdp_speed[True-False-False-False-False] 80.4500μs 19.4556μs 51.3992 KOps/s 50.2442 KOps/s $\color{#35bf28}+2.30\%$
test_step_mdp_speed[False-True-True-True-True] 93.4810μs 50.0462μs 19.9815 KOps/s 19.5474 KOps/s $\color{#35bf28}+2.22\%$
test_step_mdp_speed[False-True-True-True-False] 77.2210μs 30.8048μs 32.4624 KOps/s 31.7440 KOps/s $\color{#35bf28}+2.26\%$
test_step_mdp_speed[False-True-True-False-True] 2.3575ms 31.7591μs 31.4871 KOps/s 31.1000 KOps/s $\color{#35bf28}+1.24\%$
test_step_mdp_speed[False-True-True-False-False] 56.6000μs 18.2849μs 54.6900 KOps/s 52.7392 KOps/s $\color{#35bf28}+3.70\%$
test_step_mdp_speed[False-True-False-True-True] 96.6710μs 53.8619μs 18.5660 KOps/s 18.3074 KOps/s $\color{#35bf28}+1.41\%$
test_step_mdp_speed[False-True-False-True-False] 83.2010μs 33.7723μs 29.6101 KOps/s 29.2782 KOps/s $\color{#35bf28}+1.13\%$
test_step_mdp_speed[False-True-False-False-True] 80.5310μs 34.2226μs 29.2204 KOps/s 28.8185 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[False-True-False-False-False] 47.5210μs 20.9018μs 47.8428 KOps/s 47.0196 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[False-False-True-True-True] 0.1178ms 56.4687μs 17.7089 KOps/s 17.5592 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[False-False-True-True-False] 81.1710μs 36.7238μs 27.2303 KOps/s 27.3796 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-False-True-False-True] 73.9210μs 34.0686μs 29.3525 KOps/s 28.6010 KOps/s $\color{#35bf28}+2.63\%$
test_step_mdp_speed[False-False-True-False-False] 91.8510μs 20.5496μs 48.6627 KOps/s 46.3715 KOps/s $\color{#35bf28}+4.94\%$
test_step_mdp_speed[False-False-False-True-True] 0.1243ms 57.9040μs 17.2700 KOps/s 17.0050 KOps/s $\color{#35bf28}+1.56\%$
test_step_mdp_speed[False-False-False-True-False] 0.1034ms 38.9553μs 25.6705 KOps/s 25.3909 KOps/s $\color{#35bf28}+1.10\%$
test_step_mdp_speed[False-False-False-False-True] 73.6610μs 36.9276μs 27.0800 KOps/s 27.1529 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[False-False-False-False-False] 67.1010μs 23.4990μs 42.5550 KOps/s 41.6004 KOps/s $\color{#35bf28}+2.29\%$
test_non_tensor_env_rollout_speed[1000-single-True] 0.7458s 0.7419s 1.3479 Ops/s 1.2950 Ops/s $\color{#35bf28}+4.09\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7238s 0.6293s 1.5891 Ops/s 1.5752 Ops/s $\color{#35bf28}+0.88\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7491s 1.6732s 0.5977 Ops/s 0.5927 Ops/s $\color{#35bf28}+0.84\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5255s 1.4512s 0.6891 Ops/s 0.6841 Ops/s $\color{#35bf28}+0.74\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 1.9992s 1.9188s 0.5212 Ops/s 0.5183 Ops/s $\color{#35bf28}+0.56\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.7767s 1.6983s 0.5888 Ops/s 0.5845 Ops/s $\color{#35bf28}+0.74\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.9558s 4.7008s 0.2127 Ops/s 0.2150 Ops/s $\color{#d91a1a}-1.05\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.4923s 4.4121s 0.2266 Ops/s 0.2245 Ops/s $\color{#35bf28}+0.98\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9627s 1.8927s 0.5283 Ops/s 0.5244 Ops/s $\color{#35bf28}+0.76\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7156s 1.6296s 0.6137 Ops/s 0.6191 Ops/s $\color{#d91a1a}-0.87\%$
test_values[generalized_advantage_estimate-True-True] 10.1473ms 9.8832ms 101.1816 Ops/s 102.2118 Ops/s $\color{#d91a1a}-1.01\%$
test_values[vec_generalized_advantage_estimate-True-True] 14.9972ms 11.1656ms 89.5608 Ops/s 89.7114 Ops/s $\color{#d91a1a}-0.17\%$
test_values[td0_return_estimate-False-False] 0.2459ms 0.1272ms 7.8637 KOps/s 7.7031 KOps/s $\color{#35bf28}+2.09\%$
test_values[td1_return_estimate-False-False] 28.1828ms 27.4258ms 36.4620 Ops/s 37.8515 Ops/s $\color{#d91a1a}-3.67\%$
test_values[vec_td1_return_estimate-False-False] 11.3280ms 11.1635ms 89.5773 Ops/s 89.7575 Ops/s $\color{#d91a1a}-0.20\%$
test_values[td_lambda_return_estimate-True-False] 41.5613ms 40.1398ms 24.9129 Ops/s 25.9001 Ops/s $\color{#d91a1a}-3.81\%$
test_values[vec_td_lambda_return_estimate-True-False] 12.2813ms 11.2203ms 89.1239 Ops/s 90.2090 Ops/s $\color{#d91a1a}-1.20\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.9149ms 8.7774ms 113.9294 Ops/s 115.2713 Ops/s $\color{#d91a1a}-1.16\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.6214ms 1.4480ms 690.5875 Ops/s 680.4634 Ops/s $\color{#35bf28}+1.49\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6026ms 0.4177ms 2.3939 KOps/s 2.5154 KOps/s $\color{#d91a1a}-4.83\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 29.9452ms 29.3744ms 34.0433 Ops/s 42.7361 Ops/s $\textbf{\color{#d91a1a}-20.34\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 1.9473ms 1.7004ms 588.1124 Ops/s 587.6200 Ops/s $\color{#35bf28}+0.08\%$
test_dqn_speed[False-None] 1.5516ms 1.4019ms 713.3292 Ops/s 697.5952 Ops/s $\color{#35bf28}+2.26\%$
test_dqn_speed[False-backward] 2.0211ms 1.9007ms 526.1183 Ops/s 533.6479 Ops/s $\color{#d91a1a}-1.41\%$
test_dqn_speed[True-None] 0.9282ms 0.5301ms 1.8863 KOps/s 1.7373 KOps/s $\textbf{\color{#35bf28}+8.58\%}$
test_dqn_speed[True-backward] 1.0318ms 0.9743ms 1.0263 KOps/s 924.3203 Ops/s $\textbf{\color{#35bf28}+11.04\%}$
test_dqn_speed[reduce-overhead-None] 0.8651ms 0.5206ms 1.9209 KOps/s 1.8627 KOps/s $\color{#35bf28}+3.12\%$
test_ddpg_speed[False-None] 3.1310ms 2.7828ms 359.3553 Ops/s 356.4808 Ops/s $\color{#35bf28}+0.81\%$
test_ddpg_speed[False-backward] 4.1093ms 3.9639ms 252.2784 Ops/s 251.6219 Ops/s $\color{#35bf28}+0.26\%$
test_ddpg_speed[True-None] 1.7342ms 1.3653ms 732.4601 Ops/s 676.9427 Ops/s $\textbf{\color{#35bf28}+8.20\%}$
test_ddpg_speed[True-backward] 2.4298ms 2.3305ms 429.0896 Ops/s 405.9604 Ops/s $\textbf{\color{#35bf28}+5.70\%}$
test_ddpg_speed[reduce-overhead-None] 1.9895ms 1.3666ms 731.7522 Ops/s 713.3036 Ops/s $\color{#35bf28}+2.59\%$
test_sac_speed[False-None] 8.5337ms 7.8559ms 127.2924 Ops/s 126.3114 Ops/s $\color{#35bf28}+0.78\%$
test_sac_speed[False-backward] 11.4400ms 10.9964ms 90.9390 Ops/s 89.7003 Ops/s $\color{#35bf28}+1.38\%$
test_sac_speed[True-None] 2.2687ms 2.1186ms 472.0050 Ops/s 464.1388 Ops/s $\color{#35bf28}+1.69\%$
test_sac_speed[True-backward] 4.0345ms 3.9478ms 253.3085 Ops/s 243.4219 Ops/s $\color{#35bf28}+4.06\%$
test_sac_speed[reduce-overhead-None] 2.3899ms 2.0852ms 479.5621 Ops/s 455.6290 Ops/s $\textbf{\color{#35bf28}+5.25\%}$
test_redq_speed[False-None] 10.9026ms 10.2729ms 97.3431 Ops/s 98.6217 Ops/s $\color{#d91a1a}-1.30\%$
test_redq_speed[False-backward] 19.6204ms 18.0804ms 55.3085 Ops/s 58.2949 Ops/s $\textbf{\color{#d91a1a}-5.12\%}$
test_redq_speed[True-None] 4.8954ms 4.3636ms 229.1690 Ops/s 227.3965 Ops/s $\color{#35bf28}+0.78\%$
test_redq_speed[True-backward] 9.8191ms 9.4276ms 106.0719 Ops/s 107.5491 Ops/s $\color{#d91a1a}-1.37\%$
test_redq_speed[reduce-overhead-None] 4.6069ms 4.3120ms 231.9135 Ops/s 231.7362 Ops/s $\color{#35bf28}+0.08\%$
test_redq_deprec_speed[False-None] 11.3517ms 10.9016ms 91.7297 Ops/s 92.6837 Ops/s $\color{#d91a1a}-1.03\%$
test_redq_deprec_speed[False-backward] 0.3900s 22.9600ms 43.5539 Ops/s 64.8227 Ops/s $\textbf{\color{#d91a1a}-32.81\%}$
test_redq_deprec_speed[True-None] 4.0559ms 3.5964ms 278.0595 Ops/s 275.3466 Ops/s $\color{#35bf28}+0.99\%$
test_redq_deprec_speed[True-backward] 7.8358ms 7.4788ms 133.7115 Ops/s 128.6422 Ops/s $\color{#35bf28}+3.94\%$
test_redq_deprec_speed[reduce-overhead-None] 3.6382ms 3.4906ms 286.4832 Ops/s 275.5604 Ops/s $\color{#35bf28}+3.96\%$
test_td3_speed[False-None] 48.2118ms 8.2148ms 121.7315 Ops/s 128.4557 Ops/s $\textbf{\color{#d91a1a}-5.23\%}$
test_td3_speed[False-backward] 11.0475ms 10.5882ms 94.4450 Ops/s 93.7880 Ops/s $\color{#35bf28}+0.70\%$
test_td3_speed[True-None] 1.8531ms 1.7878ms 559.3430 Ops/s 551.7242 Ops/s $\color{#35bf28}+1.38\%$
test_td3_speed[True-backward] 3.6477ms 3.5377ms 282.6719 Ops/s 228.2323 Ops/s $\textbf{\color{#35bf28}+23.85\%}$
test_td3_speed[reduce-overhead-None] 1.7920ms 1.7413ms 574.2839 Ops/s 558.9770 Ops/s $\color{#35bf28}+2.74\%$
test_cql_speed[False-None] 28.9156ms 25.8165ms 38.7349 Ops/s 38.3649 Ops/s $\color{#35bf28}+0.96\%$
test_cql_speed[False-backward] 37.2627ms 34.4245ms 29.0491 Ops/s 28.6692 Ops/s $\color{#35bf28}+1.32\%$
test_cql_speed[True-None] 12.6271ms 12.2362ms 81.7246 Ops/s 79.4302 Ops/s $\color{#35bf28}+2.89\%$
test_cql_speed[True-backward] 18.7565ms 18.1905ms 54.9739 Ops/s 57.2654 Ops/s $\color{#d91a1a}-4.00\%$
test_cql_speed[reduce-overhead-None] 15.6203ms 12.5124ms 79.9210 Ops/s 70.2901 Ops/s $\textbf{\color{#35bf28}+13.70\%}$
test_a2c_speed[False-None] 5.6191ms 5.2718ms 189.6874 Ops/s 188.5876 Ops/s $\color{#35bf28}+0.58\%$
test_a2c_speed[False-backward] 12.1618ms 11.5967ms 86.2312 Ops/s 87.4291 Ops/s $\color{#d91a1a}-1.37\%$
test_a2c_speed[True-None] 3.9077ms 3.7306ms 268.0516 Ops/s 264.2905 Ops/s $\color{#35bf28}+1.42\%$
test_a2c_speed[True-backward] 8.7970ms 8.5186ms 117.3902 Ops/s 117.1524 Ops/s $\color{#35bf28}+0.20\%$
test_a2c_speed[reduce-overhead-None] 4.1378ms 3.7097ms 269.5642 Ops/s 271.6619 Ops/s $\color{#d91a1a}-0.77\%$
test_ppo_speed[False-None] 6.0662ms 5.7636ms 173.5016 Ops/s 174.5078 Ops/s $\color{#d91a1a}-0.58\%$
test_ppo_speed[False-backward] 12.6387ms 12.0299ms 83.1265 Ops/s 83.8270 Ops/s $\color{#d91a1a}-0.84\%$
test_ppo_speed[True-None] 3.9612ms 3.5999ms 277.7858 Ops/s 279.5345 Ops/s $\color{#d91a1a}-0.63\%$
test_ppo_speed[True-backward] 8.4999ms 8.2944ms 120.5626 Ops/s 118.1733 Ops/s $\color{#35bf28}+2.02\%$
test_ppo_speed[reduce-overhead-None] 3.7186ms 3.5903ms 278.5306 Ops/s 279.1683 Ops/s $\color{#d91a1a}-0.23\%$
test_reinforce_speed[False-None] 4.8466ms 4.4392ms 225.2636 Ops/s 222.0040 Ops/s $\color{#35bf28}+1.47\%$
test_reinforce_speed[False-backward] 7.4195ms 7.1889ms 139.1033 Ops/s 138.3660 Ops/s $\color{#35bf28}+0.53\%$
test_reinforce_speed[True-None] 3.2729ms 2.8636ms 349.2047 Ops/s 344.9362 Ops/s $\color{#35bf28}+1.24\%$
test_reinforce_speed[True-backward] 7.8991ms 7.6482ms 130.7489 Ops/s 134.4767 Ops/s $\color{#d91a1a}-2.77\%$
test_reinforce_speed[reduce-overhead-None] 3.1056ms 2.8322ms 353.0841 Ops/s 354.0306 Ops/s $\color{#d91a1a}-0.27\%$
test_iql_speed[False-None] 30.1973ms 20.7343ms 48.2292 Ops/s 50.4412 Ops/s $\color{#d91a1a}-4.39\%$
test_iql_speed[False-backward] 30.7485ms 29.8103ms 33.5454 Ops/s 33.7617 Ops/s $\color{#d91a1a}-0.64\%$
test_iql_speed[True-None] 8.6346ms 8.3823ms 119.2995 Ops/s 115.2175 Ops/s $\color{#35bf28}+3.54\%$
test_iql_speed[True-backward] 16.8586ms 16.4567ms 60.7656 Ops/s 60.2783 Ops/s $\color{#35bf28}+0.81\%$
test_iql_speed[reduce-overhead-None] 8.7986ms 8.4340ms 118.5670 Ops/s 117.2184 Ops/s $\color{#35bf28}+1.15\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1647ms 6.0316ms 165.7930 Ops/s 164.9752 Ops/s $\color{#35bf28}+0.50\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.9109ms 0.2895ms 3.4542 KOps/s 2.7721 KOps/s $\textbf{\color{#35bf28}+24.61\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6816ms 0.2621ms 3.8151 KOps/s 2.7617 KOps/s $\textbf{\color{#35bf28}+38.15\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0838ms 5.7881ms 172.7694 Ops/s 172.1444 Ops/s $\color{#35bf28}+0.36\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.9120ms 0.2714ms 3.6850 KOps/s 2.9100 KOps/s $\textbf{\color{#35bf28}+26.63\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6619ms 0.2537ms 3.9414 KOps/s 3.6657 KOps/s $\textbf{\color{#35bf28}+7.52\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5112ms 1.2705ms 787.1076 Ops/s 710.1303 Ops/s $\textbf{\color{#35bf28}+10.84\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4440ms 1.1852ms 843.7670 Ops/s 754.2579 Ops/s $\textbf{\color{#35bf28}+11.87\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1952ms 5.9493ms 168.0862 Ops/s 166.4051 Ops/s $\color{#35bf28}+1.01\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.4004ms 0.4673ms 2.1401 KOps/s 1.9530 KOps/s $\textbf{\color{#35bf28}+9.58\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6788ms 0.4489ms 2.2279 KOps/s 2.1796 KOps/s $\color{#35bf28}+2.21\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0768ms 5.8921ms 169.7202 Ops/s 171.1261 Ops/s $\color{#d91a1a}-0.82\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.0701ms 0.3387ms 2.9526 KOps/s 2.5040 KOps/s $\textbf{\color{#35bf28}+17.92\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5220ms 0.3161ms 3.1635 KOps/s 3.0107 KOps/s $\textbf{\color{#35bf28}+5.07\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.1520ms 5.7937ms 172.6009 Ops/s 172.3873 Ops/s $\color{#35bf28}+0.12\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.0756ms 0.3434ms 2.9118 KOps/s 2.9334 KOps/s $\color{#d91a1a}-0.74\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5250ms 0.3160ms 3.1645 KOps/s 3.1853 KOps/s $\color{#d91a1a}-0.65\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2148ms 5.9992ms 166.6882 Ops/s 166.3879 Ops/s $\color{#35bf28}+0.18\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.1175ms 0.5723ms 1.7472 KOps/s 2.1190 KOps/s $\textbf{\color{#d91a1a}-17.54\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7523ms 0.5603ms 1.7846 KOps/s 2.0820 KOps/s $\textbf{\color{#d91a1a}-14.28\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.5847s 16.6613ms 60.0193 Ops/s 52.0320 Ops/s $\textbf{\color{#35bf28}+15.35\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 3.9758ms 1.8197ms 549.5549 Ops/s 569.9544 Ops/s $\color{#d91a1a}-3.58\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.2115ms 1.1484ms 870.8072 Ops/s 826.5117 Ops/s $\textbf{\color{#35bf28}+5.36\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.6551ms 5.0821ms 196.7677 Ops/s 199.9187 Ops/s $\color{#d91a1a}-1.58\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 4.0280ms 1.7353ms 576.2735 Ops/s 527.1073 Ops/s $\textbf{\color{#35bf28}+9.33\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.1313ms 1.1721ms 853.1698 Ops/s 802.5789 Ops/s $\textbf{\color{#35bf28}+6.30\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.7858ms 5.1847ms 192.8748 Ops/s 192.2599 Ops/s $\color{#35bf28}+0.32\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 5.1300ms 1.9940ms 501.5078 Ops/s 484.0693 Ops/s $\color{#35bf28}+3.60\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.3202ms 1.0483ms 953.9447 Ops/s 910.1793 Ops/s $\color{#35bf28}+4.81\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 37.5448ms 35.3006ms 28.3281 Ops/s 28.0884 Ops/s $\color{#35bf28}+0.85\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 19.3372ms 17.6517ms 56.6518 Ops/s 56.4364 Ops/s $\color{#35bf28}+0.38\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 41.0252ms 37.3220ms 26.7938 Ops/s 27.1169 Ops/s $\color{#d91a1a}-1.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 19.8831ms 18.2147ms 54.9006 Ops/s 54.8967 Ops/s $+0.01\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 41.7398ms 38.4374ms 26.0163 Ops/s 25.9669 Ops/s $\color{#35bf28}+0.19\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 21.5262ms 19.7493ms 50.6347 Ops/s 50.8181 Ops/s $\color{#d91a1a}-0.36\%$
test_storage_write_lazystack[50-img_shape0-small] 0.8505ms 0.2166ms 4.6163 KOps/s 4.7135 KOps/s $\color{#d91a1a}-2.06\%$
test_storage_write_lazystack[100-img_shape1-atari] 1.6764ms 1.3619ms 734.2610 Ops/s 727.2330 Ops/s $\color{#35bf28}+0.97\%$
test_storage_write_lazystack[100-img_shape2-large_img] 2.5667ms 2.3679ms 422.3166 Ops/s 440.0158 Ops/s $\color{#d91a1a}-4.02\%$
test_storage_write_lazystack[200-img_shape3-large_batch] 3.0539ms 2.8852ms 346.5908 Ops/s 346.4085 Ops/s $\color{#35bf28}+0.05\%$
test_storage_write_contiguous[50-img_shape0-small] 0.2180ms 0.1331ms 7.5158 KOps/s 7.5964 KOps/s $\color{#d91a1a}-1.06\%$
test_storage_write_contiguous[100-img_shape1-atari] 0.3465ms 0.2131ms 4.6924 KOps/s 5.4088 KOps/s $\textbf{\color{#d91a1a}-13.24\%}$
test_storage_write_contiguous[100-img_shape2-large_img] 1.9391ms 1.7389ms 575.0700 Ops/s 569.2489 Ops/s $\color{#35bf28}+1.02\%$
test_storage_write_contiguous[200-img_shape3-large_batch] 1.5153ms 1.3288ms 752.5779 Ops/s 802.2052 Ops/s $\textbf{\color{#d91a1a}-6.19\%}$
test_collector_stack_then_write[50-img_shape0-small] 1.1985ms 1.1054ms 904.6358 Ops/s 903.6942 Ops/s $\color{#35bf28}+0.10\%$
test_collector_stack_then_write[100-img_shape1-atari] 7.5163ms 3.5436ms 282.1995 Ops/s 284.7468 Ops/s $\color{#d91a1a}-0.89\%$
test_collector_stack_then_write[100-img_shape2-large_img] 10.3715ms 5.8231ms 171.7305 Ops/s 176.3522 Ops/s $\color{#d91a1a}-2.62\%$
test_collector_stack_then_write[200-img_shape3-large_batch] 12.2680ms 6.9034ms 144.8554 Ops/s 137.9401 Ops/s $\textbf{\color{#35bf28}+5.01\%}$
test_collector_lazystack_then_write[50-img_shape0-small] 0.4351ms 0.2736ms 3.6545 KOps/s 3.5834 KOps/s $\color{#35bf28}+1.99\%$
test_collector_lazystack_then_write[100-img_shape1-atari] 1.6906ms 1.5050ms 664.4458 Ops/s 668.9916 Ops/s $\color{#d91a1a}-0.68\%$
test_collector_lazystack_then_write[100-img_shape2-large_img] 2.7815ms 2.4873ms 402.0466 Ops/s 417.9881 Ops/s $\color{#d91a1a}-3.81\%$
test_collector_lazystack_then_write[200-img_shape3-large_batch] 3.3195ms 3.0962ms 322.9790 Ops/s 319.7002 Ops/s $\color{#35bf28}+1.03\%$
test_collector_without_rb[100-img_shape0-atari] 34.8534ms 33.4738ms 29.8742 Ops/s 29.8124 Ops/s $\color{#35bf28}+0.21\%$
test_collector_without_rb[200-img_shape1-large_batch] 68.8073ms 66.7030ms 14.9918 Ops/s 15.0925 Ops/s $\color{#d91a1a}-0.67\%$
test_collector_with_rb[100-img_shape0-atari] 38.2786ms 37.7500ms 26.4901 Ops/s 25.9651 Ops/s $\color{#35bf28}+2.02\%$
test_collector_with_rb[200-img_shape1-large_batch] 96.3727ms 76.2487ms 13.1150 Ops/s 13.1920 Ops/s $\color{#d91a1a}-0.58\%$

@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

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

Total Benchmarks: 172. Improved: $\large\color{#35bf28}39$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_tensor_to_bytestream_speed[pickle] 83.0388μs 81.8126μs 12.2231 KOps/s 12.2786 KOps/s $\color{#d91a1a}-0.45\%$
test_tensor_to_bytestream_speed[torch.save] 0.1422ms 0.1417ms 7.0588 KOps/s 6.9578 KOps/s $\color{#35bf28}+1.45\%$
test_tensor_to_bytestream_speed[untyped_storage] 0.1338s 0.1337s 7.4801 Ops/s 8.6700 Ops/s $\textbf{\color{#d91a1a}-13.72\%}$
test_tensor_to_bytestream_speed[numpy] 2.8974μs 2.8888μs 346.1585 KOps/s 361.6573 KOps/s $\color{#d91a1a}-4.29\%$
test_tensor_to_bytestream_speed[safetensors] 37.6912μs 37.4866μs 26.6762 KOps/s 26.4707 KOps/s $\color{#35bf28}+0.78\%$
test_simple 0.8072s 0.8038s 1.2440 Ops/s 1.2091 Ops/s $\color{#35bf28}+2.89\%$
test_transformed 1.5625s 1.4696s 0.6805 Ops/s 0.6828 Ops/s $\color{#d91a1a}-0.34\%$
test_serial 2.4488s 2.3566s 0.4243 Ops/s 0.4231 Ops/s $\color{#35bf28}+0.30\%$
test_parallel 1.9227s 1.8235s 0.5484 Ops/s 0.5423 Ops/s $\color{#35bf28}+1.12\%$
test_step_mdp_speed[True-True-True-True-True] 0.2560ms 44.1103μs 22.6705 KOps/s 19.8514 KOps/s $\textbf{\color{#35bf28}+14.20\%}$
test_step_mdp_speed[True-True-True-True-False] 58.7330μs 25.2131μs 39.6620 KOps/s 36.5901 KOps/s $\textbf{\color{#35bf28}+8.40\%}$
test_step_mdp_speed[True-True-True-False-True] 59.8130μs 24.7112μs 40.4674 KOps/s 36.4086 KOps/s $\textbf{\color{#35bf28}+11.15\%}$
test_step_mdp_speed[True-True-True-False-False] 48.5720μs 13.8866μs 72.0116 KOps/s 66.4210 KOps/s $\textbf{\color{#35bf28}+8.42\%}$
test_step_mdp_speed[True-True-False-True-True] 95.8940μs 47.5981μs 21.0093 KOps/s 18.7769 KOps/s $\textbf{\color{#35bf28}+11.89\%}$
test_step_mdp_speed[True-True-False-True-False] 69.8130μs 27.7209μs 36.0739 KOps/s 31.4395 KOps/s $\textbf{\color{#35bf28}+14.74\%}$
test_step_mdp_speed[True-True-False-False-True] 59.6530μs 27.4296μs 36.4569 KOps/s 32.0432 KOps/s $\textbf{\color{#35bf28}+13.77\%}$
test_step_mdp_speed[True-True-False-False-False] 77.0040μs 16.6658μs 60.0031 KOps/s 53.7444 KOps/s $\textbf{\color{#35bf28}+11.65\%}$
test_step_mdp_speed[True-False-True-True-True] 99.7850μs 50.7115μs 19.7194 KOps/s 17.4839 KOps/s $\textbf{\color{#35bf28}+12.79\%}$
test_step_mdp_speed[True-False-True-True-False] 71.7630μs 30.6323μs 32.6453 KOps/s 28.9841 KOps/s $\textbf{\color{#35bf28}+12.63\%}$
test_step_mdp_speed[True-False-True-False-True] 57.2030μs 27.7646μs 36.0170 KOps/s 32.9081 KOps/s $\textbf{\color{#35bf28}+9.45\%}$
test_step_mdp_speed[True-False-True-False-False] 97.4240μs 16.5634μs 60.3741 KOps/s 54.1486 KOps/s $\textbf{\color{#35bf28}+11.50\%}$
test_step_mdp_speed[True-False-False-True-True] 93.2850μs 53.0285μs 18.8578 KOps/s 17.0308 KOps/s $\textbf{\color{#35bf28}+10.73\%}$
test_step_mdp_speed[True-False-False-True-False] 74.9430μs 33.0847μs 30.2255 KOps/s 27.8604 KOps/s $\textbf{\color{#35bf28}+8.49\%}$
test_step_mdp_speed[True-False-False-False-True] 86.7340μs 30.5406μs 32.7433 KOps/s 30.4321 KOps/s $\textbf{\color{#35bf28}+7.59\%}$
test_step_mdp_speed[True-False-False-False-False] 0.1251ms 18.7745μs 53.2637 KOps/s 47.3954 KOps/s $\textbf{\color{#35bf28}+12.38\%}$
test_step_mdp_speed[False-True-True-True-True] 84.8940μs 50.4805μs 19.8096 KOps/s 17.9353 KOps/s $\textbf{\color{#35bf28}+10.45\%}$
test_step_mdp_speed[False-True-True-True-False] 76.9830μs 30.3750μs 32.9218 KOps/s 28.1637 KOps/s $\textbf{\color{#35bf28}+16.89\%}$
test_step_mdp_speed[False-True-True-False-True] 2.3485ms 31.6645μs 31.5811 KOps/s 31.3167 KOps/s $\color{#35bf28}+0.84\%$
test_step_mdp_speed[False-True-True-False-False] 49.8430μs 18.5211μs 53.9924 KOps/s 53.9469 KOps/s $\color{#35bf28}+0.08\%$
test_step_mdp_speed[False-True-False-True-True] 94.0540μs 52.5682μs 19.0229 KOps/s 18.5946 KOps/s $\color{#35bf28}+2.30\%$
test_step_mdp_speed[False-True-False-True-False] 71.7040μs 33.6122μs 29.7511 KOps/s 29.7414 KOps/s $\color{#35bf28}+0.03\%$
test_step_mdp_speed[False-True-False-False-True] 67.6830μs 33.3800μs 29.9581 KOps/s 26.1401 KOps/s $\textbf{\color{#35bf28}+14.61\%}$
test_step_mdp_speed[False-True-False-False-False] 54.1330μs 21.2483μs 47.0626 KOps/s 41.0055 KOps/s $\textbf{\color{#35bf28}+14.77\%}$
test_step_mdp_speed[False-False-True-True-True] 94.4750μs 56.6531μs 17.6513 KOps/s 15.6917 KOps/s $\textbf{\color{#35bf28}+12.49\%}$
test_step_mdp_speed[False-False-True-True-False] 69.6930μs 36.1498μs 27.6627 KOps/s 24.6688 KOps/s $\textbf{\color{#35bf28}+12.14\%}$
test_step_mdp_speed[False-False-True-False-True] 72.2730μs 34.0977μs 29.3275 KOps/s 26.2611 KOps/s $\textbf{\color{#35bf28}+11.68\%}$
test_step_mdp_speed[False-False-True-False-False] 53.5830μs 21.0182μs 47.5778 KOps/s 45.3820 KOps/s $\color{#35bf28}+4.84\%$
test_step_mdp_speed[False-False-False-True-True] 86.9240μs 59.1522μs 16.9055 KOps/s 16.4278 KOps/s $\color{#35bf28}+2.91\%$
test_step_mdp_speed[False-False-False-True-False] 69.8130μs 38.9971μs 25.6429 KOps/s 22.6971 KOps/s $\textbf{\color{#35bf28}+12.98\%}$
test_step_mdp_speed[False-False-False-False-True] 72.3830μs 36.3648μs 27.4991 KOps/s 24.9705 KOps/s $\textbf{\color{#35bf28}+10.13\%}$
test_step_mdp_speed[False-False-False-False-False] 59.6930μs 23.7286μs 42.1433 KOps/s 37.1023 KOps/s $\textbf{\color{#35bf28}+13.59\%}$
test_non_tensor_env_rollout_speed[1000-single-True] 0.8825s 0.7779s 1.2855 Ops/s 1.2588 Ops/s $\color{#35bf28}+2.12\%$
test_non_tensor_env_rollout_speed[1000-single-False] 0.7441s 0.6397s 1.5632 Ops/s 1.5230 Ops/s $\color{#35bf28}+2.64\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-True] 1.7775s 1.7037s 0.5869 Ops/s 0.5823 Ops/s $\color{#35bf28}+0.80\%$
test_non_tensor_env_rollout_speed[1000-serial-no-buffers-False] 1.5569s 1.4720s 0.6794 Ops/s 0.6788 Ops/s $\color{#35bf28}+0.09\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-True] 2.0266s 1.9507s 0.5126 Ops/s 0.5056 Ops/s $\color{#35bf28}+1.39\%$
test_non_tensor_env_rollout_speed[1000-serial-buffers-False] 1.8037s 1.7237s 0.5801 Ops/s 0.5730 Ops/s $\color{#35bf28}+1.24\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-True] 4.8635s 4.7661s 0.2098 Ops/s 0.2125 Ops/s $\color{#d91a1a}-1.28\%$
test_non_tensor_env_rollout_speed[1000-parallel-no-buffers-False] 4.7567s 4.5417s 0.2202 Ops/s 0.2183 Ops/s $\color{#35bf28}+0.85\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-True] 1.9625s 1.9036s 0.5253 Ops/s 0.5237 Ops/s $\color{#35bf28}+0.31\%$
test_non_tensor_env_rollout_speed[1000-parallel-buffers-False] 1.7939s 1.6542s 0.6045 Ops/s 0.6098 Ops/s $\color{#d91a1a}-0.86\%$
test_values[generalized_advantage_estimate-True-True] 22.0671ms 21.5249ms 46.4579 Ops/s 47.2714 Ops/s $\color{#d91a1a}-1.72\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1439s 3.8207ms 261.7287 Ops/s 263.6473 Ops/s $\color{#d91a1a}-0.73\%$
test_values[td0_return_estimate-False-False] 0.1100ms 88.3533μs 11.3182 KOps/s 11.6958 KOps/s $\color{#d91a1a}-3.23\%$
test_values[td1_return_estimate-False-False] 52.7165ms 51.7716ms 19.3156 Ops/s 19.7682 Ops/s $\color{#d91a1a}-2.29\%$
test_values[vec_td1_return_estimate-False-False] 1.3796ms 1.1168ms 895.4446 Ops/s 904.3558 Ops/s $\color{#d91a1a}-0.99\%$
test_values[td_lambda_return_estimate-True-False] 87.3997ms 85.6242ms 11.6789 Ops/s 12.0716 Ops/s $\color{#d91a1a}-3.25\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.3180ms 1.1123ms 899.0601 Ops/s 904.1339 Ops/s $\color{#d91a1a}-0.56\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 22.7414ms 21.8242ms 45.8207 Ops/s 45.4966 Ops/s $\color{#35bf28}+0.71\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0537ms 0.8148ms 1.2272 KOps/s 1.2641 KOps/s $\color{#d91a1a}-2.91\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8427ms 0.7156ms 1.3974 KOps/s 1.4243 KOps/s $\color{#d91a1a}-1.89\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5812ms 1.5251ms 655.6751 Ops/s 658.6620 Ops/s $\color{#d91a1a}-0.45\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7857ms 0.7307ms 1.3686 KOps/s 1.3698 KOps/s $\color{#d91a1a}-0.08\%$
test_dqn_speed[False-None] 1.7099ms 1.5843ms 631.1810 Ops/s 618.0478 Ops/s $\color{#35bf28}+2.12\%$
test_dqn_speed[False-backward] 2.3367ms 2.2246ms 449.5198 Ops/s 445.8762 Ops/s $\color{#35bf28}+0.82\%$
test_dqn_speed[True-None] 0.6371ms 0.5717ms 1.7492 KOps/s 1.7513 KOps/s $\color{#d91a1a}-0.12\%$
test_dqn_speed[True-backward] 1.2865ms 1.2263ms 815.4885 Ops/s 826.2168 Ops/s $\color{#d91a1a}-1.30\%$
test_dqn_speed[reduce-overhead-None] 0.6942ms 0.5937ms 1.6844 KOps/s 1.6501 KOps/s $\color{#35bf28}+2.08\%$
test_ddpg_speed[False-None] 3.2869ms 2.9463ms 339.4127 Ops/s 339.3226 Ops/s $\color{#35bf28}+0.03\%$
test_ddpg_speed[False-backward] 4.6698ms 4.3762ms 228.5096 Ops/s 225.6198 Ops/s $\color{#35bf28}+1.28\%$
test_ddpg_speed[True-None] 1.6131ms 1.3405ms 746.0026 Ops/s 745.1890 Ops/s $\color{#35bf28}+0.11\%$
test_ddpg_speed[True-backward] 2.6217ms 2.5626ms 390.2345 Ops/s 394.0077 Ops/s $\color{#d91a1a}-0.96\%$
test_ddpg_speed[reduce-overhead-None] 1.5335ms 1.3677ms 731.1449 Ops/s 737.9589 Ops/s $\color{#d91a1a}-0.92\%$
test_sac_speed[False-None] 9.1014ms 8.5517ms 116.9356 Ops/s 117.0353 Ops/s $\color{#d91a1a}-0.09\%$
test_sac_speed[False-backward] 12.3771ms 11.7898ms 84.8192 Ops/s 84.0434 Ops/s $\color{#35bf28}+0.92\%$
test_sac_speed[True-None] 1.9582ms 1.8597ms 537.7282 Ops/s 536.9506 Ops/s $\color{#35bf28}+0.14\%$
test_sac_speed[True-backward] 3.6905ms 3.5574ms 281.1060 Ops/s 284.9320 Ops/s $\color{#d91a1a}-1.34\%$
test_sac_speed[reduce-overhead-None] 0.3603s 11.9718ms 83.5295 Ops/s 91.6937 Ops/s $\textbf{\color{#d91a1a}-8.90\%}$
test_redq_deprec_speed[False-None] 9.8336ms 9.5294ms 104.9380 Ops/s 104.8243 Ops/s $\color{#35bf28}+0.11\%$
test_redq_deprec_speed[False-backward] 13.4838ms 12.7927ms 78.1699 Ops/s 78.4903 Ops/s $\color{#d91a1a}-0.41\%$
test_redq_deprec_speed[True-None] 2.7010ms 2.6047ms 383.9191 Ops/s 391.9532 Ops/s $\color{#d91a1a}-2.05\%$
test_redq_deprec_speed[True-backward] 4.6876ms 4.2654ms 234.4451 Ops/s 235.3606 Ops/s $\color{#d91a1a}-0.39\%$
test_redq_deprec_speed[reduce-overhead-None] 15.7476ms 9.7607ms 102.4514 Ops/s 101.3899 Ops/s $\color{#35bf28}+1.05\%$
test_td3_speed[False-None] 8.4910ms 8.3778ms 119.3635 Ops/s 117.0279 Ops/s $\color{#35bf28}+2.00\%$
test_td3_speed[False-backward] 11.3563ms 10.8172ms 92.4458 Ops/s 91.4755 Ops/s $\color{#35bf28}+1.06\%$
test_td3_speed[True-None] 1.7166ms 1.6950ms 589.9618 Ops/s 601.5322 Ops/s $\color{#d91a1a}-1.92\%$
test_td3_speed[True-backward] 3.2822ms 3.1796ms 314.5078 Ops/s 299.0932 Ops/s $\textbf{\color{#35bf28}+5.15\%}$
test_td3_speed[reduce-overhead-None] 46.2858ms 23.8145ms 41.9913 Ops/s 39.5782 Ops/s $\textbf{\color{#35bf28}+6.10\%}$
test_cql_speed[False-None] 18.1900ms 17.6474ms 56.6656 Ops/s 56.5301 Ops/s $\color{#35bf28}+0.24\%$
test_cql_speed[False-backward] 23.7764ms 22.9595ms 43.5550 Ops/s 42.4470 Ops/s $\color{#35bf28}+2.61\%$
test_cql_speed[True-None] 3.5562ms 3.3266ms 300.6035 Ops/s 303.3208 Ops/s $\color{#d91a1a}-0.90\%$
test_cql_speed[True-backward] 5.9437ms 5.5037ms 181.6943 Ops/s 179.0186 Ops/s $\color{#35bf28}+1.49\%$
test_cql_speed[reduce-overhead-None] 18.9549ms 11.9916ms 83.3914 Ops/s 83.6353 Ops/s $\color{#d91a1a}-0.29\%$
test_a2c_speed[False-None] 4.1406ms 3.3442ms 299.0289 Ops/s 297.8898 Ops/s $\color{#35bf28}+0.38\%$
test_a2c_speed[False-backward] 6.7899ms 6.3056ms 158.5893 Ops/s 151.2504 Ops/s $\color{#35bf28}+4.85\%$
test_a2c_speed[True-None] 1.5279ms 1.3650ms 732.5958 Ops/s 730.6020 Ops/s $\color{#35bf28}+0.27\%$
test_a2c_speed[True-backward] 3.1657ms 3.0357ms 329.4135 Ops/s 332.8588 Ops/s $\color{#d91a1a}-1.04\%$
test_a2c_speed[reduce-overhead-None] 1.0910ms 1.0022ms 997.8370 Ops/s 1.0004 KOps/s $\color{#d91a1a}-0.25\%$
test_ppo_speed[False-None] 4.1318ms 3.9605ms 252.4943 Ops/s 254.3178 Ops/s $\color{#d91a1a}-0.72\%$
test_ppo_speed[False-backward] 7.6301ms 7.1385ms 140.0860 Ops/s 140.5960 Ops/s $\color{#d91a1a}-0.36\%$
test_ppo_speed[True-None] 1.5684ms 1.4578ms 685.9530 Ops/s 689.9405 Ops/s $\color{#d91a1a}-0.58\%$
test_ppo_speed[True-backward] 3.2585ms 3.1535ms 317.1076 Ops/s 303.8535 Ops/s $\color{#35bf28}+4.36\%$
test_ppo_speed[reduce-overhead-None] 1.2828ms 1.0645ms 939.4309 Ops/s 917.7099 Ops/s $\color{#35bf28}+2.37\%$
test_reinforce_speed[False-None] 2.4375ms 2.3444ms 426.5447 Ops/s 430.3420 Ops/s $\color{#d91a1a}-0.88\%$
test_reinforce_speed[False-backward] 3.4997ms 3.3864ms 295.2963 Ops/s 285.9287 Ops/s $\color{#35bf28}+3.28\%$
test_reinforce_speed[True-None] 1.4076ms 1.3116ms 762.4457 Ops/s 778.7481 Ops/s $\color{#d91a1a}-2.09\%$
test_reinforce_speed[True-backward] 3.0219ms 2.9466ms 339.3787 Ops/s 334.1502 Ops/s $\color{#35bf28}+1.56\%$
test_reinforce_speed[reduce-overhead-None] 17.4928ms 9.4766ms 105.5227 Ops/s 106.1754 Ops/s $\color{#d91a1a}-0.61\%$
test_iql_speed[False-None] 10.0940ms 9.6065ms 104.0964 Ops/s 103.7689 Ops/s $\color{#35bf28}+0.32\%$
test_iql_speed[False-backward] 14.0404ms 13.3628ms 74.8347 Ops/s 74.2035 Ops/s $\color{#35bf28}+0.85\%$
test_iql_speed[True-None] 2.4516ms 2.2289ms 448.6598 Ops/s 452.0866 Ops/s $\color{#d91a1a}-0.76\%$
test_iql_speed[True-backward] 5.2997ms 4.8442ms 206.4315 Ops/s 201.0537 Ops/s $\color{#35bf28}+2.67\%$
test_iql_speed[reduce-overhead-None] 17.7442ms 10.4897ms 95.3317 Ops/s 74.4859 Ops/s $\textbf{\color{#35bf28}+27.99\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.1469ms 6.0114ms 166.3515 Ops/s 163.0877 Ops/s $\color{#35bf28}+2.00\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.7160ms 0.2885ms 3.4660 KOps/s 2.7202 KOps/s $\textbf{\color{#35bf28}+27.42\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5155ms 0.2704ms 3.6986 KOps/s 3.1971 KOps/s $\textbf{\color{#35bf28}+15.69\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0579ms 5.8078ms 172.1824 Ops/s 167.2404 Ops/s $\color{#35bf28}+2.96\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5984ms 0.3657ms 2.7344 KOps/s 2.9732 KOps/s $\textbf{\color{#d91a1a}-8.03\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5823ms 0.3489ms 2.8665 KOps/s 3.0757 KOps/s $\textbf{\color{#d91a1a}-6.80\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.8026ms 1.4934ms 669.6192 Ops/s 669.4272 Ops/s $\color{#35bf28}+0.03\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.4792ms 1.2423ms 804.9898 Ops/s 744.6841 Ops/s $\textbf{\color{#35bf28}+8.10\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.1847ms 5.9904ms 166.9351 Ops/s 161.8471 Ops/s $\color{#35bf28}+3.14\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7439ms 0.4333ms 2.3076 KOps/s 2.0855 KOps/s $\textbf{\color{#35bf28}+10.65\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6307ms 0.4234ms 2.3621 KOps/s 2.1877 KOps/s $\textbf{\color{#35bf28}+7.97\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.9723ms 5.8327ms 171.4468 Ops/s 166.2396 Ops/s $\color{#35bf28}+3.13\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.3051ms 0.3753ms 2.6644 KOps/s 3.0855 KOps/s $\textbf{\color{#d91a1a}-13.65\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5289ms 0.3500ms 2.8575 KOps/s 3.2403 KOps/s $\textbf{\color{#d91a1a}-11.81\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.0894ms 5.8458ms 171.0621 Ops/s 166.9974 Ops/s $\color{#35bf28}+2.43\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5486ms 0.2844ms 3.5163 KOps/s 3.1599 KOps/s $\textbf{\color{#35bf28}+11.28\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.4791ms 0.2665ms 3.7518 KOps/s 2.8466 KOps/s $\textbf{\color{#35bf28}+31.80\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2284ms 6.0329ms 165.7573 Ops/s 161.6091 Ops/s $\color{#35bf28}+2.57\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9174ms 0.5060ms 1.9763 KOps/s 2.1891 KOps/s $\textbf{\color{#d91a1a}-9.72\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7483ms 0.5339ms 1.8730 KOps/s 2.3582 KOps/s $\textbf{\color{#d91a1a}-20.58\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.7106ms 5.1576ms 193.8872 Ops/s 51.3213 Ops/s $\textbf{\color{#35bf28}+277.79\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 12.4783ms 2.1370ms 467.9351 Ops/s 524.1439 Ops/s $\textbf{\color{#d91a1a}-10.72\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.2486ms 0.9535ms 1.0487 KOps/s 1.0710 KOps/s $\color{#d91a1a}-2.08\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 6.5202ms 5.1018ms 196.0104 Ops/s 192.3901 Ops/s $\color{#35bf28}+1.88\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.2200ms 1.9555ms 511.3851 Ops/s 527.3857 Ops/s $\color{#d91a1a}-3.03\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.1143ms 0.9610ms 1.0406 KOps/s 1.0580 KOps/s $\color{#d91a1a}-1.65\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.5866s 17.0930ms 58.5036 Ops/s 184.4567 Ops/s $\textbf{\color{#d91a1a}-68.28\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 8.0936ms 2.1949ms 455.6071 Ops/s 464.5042 Ops/s $\color{#d91a1a}-1.92\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0923ms 1.1690ms 855.4253 Ops/s 792.0671 Ops/s $\textbf{\color{#35bf28}+8.00\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 39.6203ms 36.8592ms 27.1302 Ops/s 27.0876 Ops/s $\color{#35bf28}+0.16\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.6007ms 18.8174ms 53.1423 Ops/s 54.0489 Ops/s $\color{#d91a1a}-1.68\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 42.0498ms 38.5956ms 25.9097 Ops/s 26.1565 Ops/s $\color{#d91a1a}-0.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 20.9380ms 19.0225ms 52.5694 Ops/s 52.7079 Ops/s $\color{#d91a1a}-0.26\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 46.6483ms 40.2086ms 24.8703 Ops/s 25.0461 Ops/s $\color{#d91a1a}-0.70\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 22.6790ms 21.0833ms 47.4309 Ops/s 48.7666 Ops/s $\color{#d91a1a}-2.74\%$
test_storage_write_lazystack[50-img_shape0-small] 0.9075ms 0.2209ms 4.5263 KOps/s 4.4020 KOps/s $\color{#35bf28}+2.82\%$
test_storage_write_lazystack[100-img_shape1-atari] 1.6400ms 1.4083ms 710.0543 Ops/s 714.1538 Ops/s $\color{#d91a1a}-0.57\%$
test_storage_write_lazystack[100-img_shape2-large_img] 2.7588ms 2.3454ms 426.3754 Ops/s 457.2268 Ops/s $\textbf{\color{#d91a1a}-6.75\%}$
test_storage_write_lazystack[200-img_shape3-large_batch] 3.0972ms 2.9133ms 343.2532 Ops/s 345.9253 Ops/s $\color{#d91a1a}-0.77\%$
test_storage_write_contiguous[50-img_shape0-small] 0.2473ms 0.1643ms 6.0876 KOps/s 5.9786 KOps/s $\color{#35bf28}+1.82\%$
test_storage_write_contiguous[100-img_shape1-atari] 0.3912ms 0.2320ms 4.3102 KOps/s 3.6321 KOps/s $\textbf{\color{#35bf28}+18.67\%}$
test_storage_write_contiguous[100-img_shape2-large_img] 1.9508ms 1.8203ms 549.3561 Ops/s 557.9625 Ops/s $\color{#d91a1a}-1.54\%$
test_storage_write_contiguous[200-img_shape3-large_batch] 1.4857ms 1.3311ms 751.2384 Ops/s 737.2751 Ops/s $\color{#35bf28}+1.89\%$
test_collector_stack_then_write[50-img_shape0-small] 1.3448ms 1.1499ms 869.6224 Ops/s 861.7653 Ops/s $\color{#35bf28}+0.91\%$
test_collector_stack_then_write[100-img_shape1-atari] 7.6084ms 3.6283ms 275.6083 Ops/s 276.5432 Ops/s $\color{#d91a1a}-0.34\%$
test_collector_stack_then_write[100-img_shape2-large_img] 8.8517ms 5.8743ms 170.2325 Ops/s 171.9084 Ops/s $\color{#d91a1a}-0.97\%$
test_collector_stack_then_write[200-img_shape3-large_batch] 7.5109ms 7.3095ms 136.8082 Ops/s 138.8588 Ops/s $\color{#d91a1a}-1.48\%$
test_collector_lazystack_then_write[50-img_shape0-small] 0.4383ms 0.2753ms 3.6330 KOps/s 3.5624 KOps/s $\color{#35bf28}+1.98\%$
test_collector_lazystack_then_write[100-img_shape1-atari] 1.6708ms 1.5300ms 653.6120 Ops/s 676.2391 Ops/s $\color{#d91a1a}-3.35\%$
test_collector_lazystack_then_write[100-img_shape2-large_img] 2.5662ms 2.4090ms 415.1178 Ops/s 430.4247 Ops/s $\color{#d91a1a}-3.56\%$
test_collector_lazystack_then_write[200-img_shape3-large_batch] 3.4248ms 3.1118ms 321.3598 Ops/s 321.6092 Ops/s $\color{#d91a1a}-0.08\%$
test_collector_without_rb[100-img_shape0-atari] 35.1289ms 34.6790ms 28.8359 Ops/s 28.8207 Ops/s $\color{#35bf28}+0.05\%$
test_collector_without_rb[200-img_shape1-large_batch] 68.3118ms 68.0585ms 14.6932 Ops/s 14.6115 Ops/s $\color{#35bf28}+0.56\%$
test_collector_with_rb[100-img_shape0-atari] 39.8640ms 39.2998ms 25.4454 Ops/s 25.3551 Ops/s $\color{#35bf28}+0.36\%$
test_collector_with_rb[200-img_shape1-large_batch] 79.3860ms 76.7441ms 13.0303 Ops/s 12.8393 Ops/s $\color{#35bf28}+1.49\%$
test_collector_without_rb_cuda[100-img_shape0-atari] 60.1702ms 57.6252ms 17.3535 Ops/s 17.3087 Ops/s $\color{#35bf28}+0.26\%$
test_collector_without_rb_cuda[200-img_shape1-large_batch] 0.8190s 0.1928s 5.1875 Ops/s 8.6664 Ops/s $\textbf{\color{#d91a1a}-40.14\%}$
test_collector_with_rb_cuda[100-img_shape0-atari] 60.0857ms 59.6567ms 16.7626 Ops/s 16.6427 Ops/s $\color{#35bf28}+0.72\%$
test_collector_with_rb_cuda[200-img_shape1-large_batch] 0.1192s 0.1186s 8.4315 Ops/s 8.3809 Ops/s $\color{#35bf28}+0.60\%$

@vmoens vmoens mentioned this pull request Feb 10, 2026
Closed
@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

⚠️ PR Title Label Error

Unknown or invalid prefix [Example].

Current title: [Example] Dreamer: add IsaacLab training script and config

Supported Prefixes (case-sensitive)

Your PR title must start with exactly one of these prefixes:

Prefix Label Applied Example
[BugFix] BugFix [BugFix] Fix memory leak in collector
[Feature] Feature [Feature] Add new optimizer
[Doc] or [Docs] Documentation [Doc] Update installation guide
[Refactor] Refactoring [Refactor] Clean up module imports
[CI] CI [CI] Fix workflow permissions
[Test] or [Tests] Tests [Tests] Add unit tests for buffer
[Environment] or [Environments] Environments [Environments] Add Gymnasium support
[Data] Data [Data] Fix replay buffer sampling
[Performance] or [Perf] Performance [Performance] Optimize tensor ops
[BC-Breaking] bc breaking [BC-Breaking] Remove deprecated API
[Deprecation] Deprecation [Deprecation] Mark old function

Note: Common variations like singular/plural are supported (e.g., [Doc] or [Docs]).

@vmoens vmoens mentioned this pull request Feb 10, 2026
Closed
@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

⚠️ PR Title Label Error

Unknown or invalid prefix [Example].

Current title: [Example] Dreamer: add IsaacLab training script and config

Supported Prefixes (case-sensitive)

Your PR title must start with exactly one of these prefixes:

Prefix Label Applied Example
[BugFix] BugFix [BugFix] Fix memory leak in collector
[Feature] Feature [Feature] Add new optimizer
[Doc] or [Docs] Documentation [Doc] Update installation guide
[Refactor] Refactoring [Refactor] Clean up module imports
[CI] CI [CI] Fix workflow permissions
[Test] or [Tests] Tests [Tests] Add unit tests for buffer
[Environment] or [Environments] Environments [Environments] Add Gymnasium support
[Data] Data [Data] Fix replay buffer sampling
[Performance] or [Perf] Performance [Performance] Optimize tensor ops
[BC-Breaking] bc breaking [BC-Breaking] Remove deprecated API
[Deprecation] Deprecation [Deprecation] Mark old function

Note: Common variations like singular/plural are supported (e.g., [Doc] or [Docs]).

@vmoens vmoens mentioned this pull request Feb 10, 2026
Closed
@github-actions
Copy link
Contributor

github-actions bot commented Feb 10, 2026

⚠️ PR Title Label Error

Unknown or invalid prefix [Example].

Current title: [Example] Dreamer: add IsaacLab training script and config

Supported Prefixes (case-sensitive)

Your PR title must start with exactly one of these prefixes:

Prefix Label Applied Example
[BugFix] BugFix [BugFix] Fix memory leak in collector
[Feature] Feature [Feature] Add new optimizer
[Doc] or [Docs] Documentation [Doc] Update installation guide
[Refactor] Refactoring [Refactor] Clean up module imports
[CI] CI [CI] Fix workflow permissions
[Test] or [Tests] Tests [Tests] Add unit tests for buffer
[Environment] or [Environments] Environments [Environments] Add Gymnasium support
[Data] Data [Data] Fix replay buffer sampling
[Performance] or [Perf] Performance [Performance] Optimize tensor ops
[BC-Breaking] bc breaking [BC-Breaking] Remove deprecated API
[Deprecation] Deprecation [Deprecation] Mark old function

Note: Common variations like singular/plural are supported (e.g., [Doc] or [Docs]).

@vmoens
Copy link
Collaborator Author

vmoens commented Feb 10, 2026

Superseded by rebuilt stack (cleanup folded into the right commits)

@vmoens vmoens closed this Feb 10, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. sota-implementations/

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@vmoens