8e77625cf1
clean
2025-11-19 16:22:01 +00:00
514e3cd83d
add convert scripts
2025-11-19 16:11:07 +00:00
25ef3c520c
fix test, add to train
2025-11-18 23:58:27 +01:00
464f14a0f4
fix imports
2025-11-18 18:41:01 +01:00
61f8c8a0f2
add docs
2025-11-18 18:37:27 +01:00
b196f04d48
more changes
2025-11-18 18:29:19 +01:00
91768a9879
Merge branch 'main' into refactor/env-libero
2025-11-18 17:43:48 +01:00
4fb41d3e5a
style
2025-11-18 17:29:25 +01:00
b257e02ccf
add env processor
2025-11-18 17:27:43 +01:00
b464d9f8bc
Fix episode filtering bug when requesting a subset of the episodes in a dataset ( #2456 )
...
* filter episodes in load_nested_dataset
* nit
* remove test filtering
* move import to module level
* added missing episode indices to the EpisodeAwareSampler in lerobot_train.py;
2025-11-18 17:26:41 +01:00
8915c6cd25
more changes
2025-11-18 16:28:18 +01:00
0ed2f87fba
iterate on review:
2025-11-18 16:12:09 +01:00
a068618faf
more styling fixes:
2025-11-18 16:01:41 +01:00
769eb27c87
more fixes
2025-11-18 15:54:14 +01:00
c3b8f65a8c
put axis-1
2025-11-18 15:40:23 +01:00
33a8d31af0
Merge branch 'main' into refactor/env-libero
2025-11-18 15:33:45 +01:00
6c9f169996
clean
2025-11-18 15:26:12 +01:00
9979b62c52
more
2025-11-18 14:57:43 +01:00
e91e48b79c
fix style
2025-11-18 14:25:48 +01:00
b4b5d057b1
more fixes
2025-11-18 14:24:59 +01:00
9a115c303c
more changes
2025-11-18 14:13:17 +01:00
e5efb6b6dc
working changes
2025-11-18 13:55:45 +01:00
4c67330430
more changes
2025-11-18 10:33:23 +01:00
784cdae55a
Fixes in port droid scripts ( #2455 )
...
* Fixes in port droid scripts
* revert default mem-per-cpu
* style nit
* fix relative imports
* style nit
2025-11-17 23:42:30 +01:00
d9e74a9d37
chore(dependencies): Bump lerobot to 0.4.2 ( #2423 )
2025-11-12 13:13:57 +01:00
a5b29d4301
chore(installation): remove libero installation patch ( #2416 )
...
* chore(installation): remove libero installation patch
* fix(ci): exclude groot for unbound deps test
2025-11-10 11:51:52 +01:00
a4aa316470
fix(dataset): fix data access bottleneck for faster training ( #2408 )
2025-11-07 21:54:44 +01:00
f6b16f6d97
fix(dataset_tools) Critical bug in modify features ( #2342 )
...
* fix bug in `_copy_data_with_feature_changes`
* Update src/lerobot/datasets/dataset_tools.py
Co-authored-by: Caroline Pascal <caroline8.pascal@gmail.com >
Signed-off-by: Michel Aractingi <michel.aractingi@huggingface.co >
* add missing import
---------
Signed-off-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Caroline Pascal <caroline8.pascal@gmail.com >
2025-11-04 15:56:41 +01:00
df0c335a5a
feat(sim): EnvHub - allow loading envs from the hub ( #2121 )
...
* add env from the hub support
* add safe loading
* changes
* add tests, docs
* more
* style/cleaning
* order
---------
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
2025-11-04 14:52:46 +01:00
87ed3a2b6e
dep(upgrade): add libero as a pypi package ( #2365 )
...
* add changes
* Update pyproject.toml
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
* add openpi-transformers
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
* new changes
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
* Update hf-libero version in pyproject.toml
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
---------
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-11-04 10:43:52 +01:00
d57d1aa197
fix(make_policy): rename mapping edge cases in training ( #2332 )
...
* fix bug
* update fixes
* add hf license
* more fixes
* add transformers
* iterate on review
* more fixes
* more fixes
* add a False test
* reduce img size
* reduce img size
* skip the test
* add
* add style
2025-10-31 13:08:42 +01:00
3f8c5d9809
fix(video_key typo): fixing video_key typo in update_video_info ( #2323 )
2025-10-28 09:41:33 +01:00
d1548e1d13
docs(install): imrpove groot and libero installation instructions ( #2314 )
2025-10-26 15:37:41 +08:00
d11ec6b5ef
docs(readme): update installation instructions for 0.4.0 ( #2310 )
2025-10-24 17:31:37 +02:00
c75455a6de
chore(dependecies): Bump lerobot to 0.4.1 ( #2299 )
...
Signed-off-by: Steven Palma <imstevenpmwork@ieee.org >
2025-10-23 20:59:30 +02:00
f25ac02e6c
chore(dependencies): Bump lerobot to 0.4.0 ( #2298 )
...
Signed-off-by: Steven Palma <imstevenpmwork@ieee.org >
2025-10-23 20:20:52 +02:00
23cb668cac
fix(ci): add fastapi dep + bump to 0.3.5 ( #2301 )
2025-10-23 19:53:44 +02:00
2ea3043b1b
patch(ci): remove pi & libero tags from PyPi release temporary due to their reliance on git dependencies ( #2300 )
2025-10-23 19:37:11 +02:00
0f61e2415f
chore(deps): update requirements file ( #2297 )
2025-10-23 18:38:41 +02:00
76a425c600
Fix: check_cached_episodes doesn't check if the requested episode video were downloaded ( #2296 )
...
* In `check_cached_episodes_sufficient` check whether all the requested video files are downloaded
* optimize loop over the video paths
* revert example num_workers
* Apply suggestion from @Copilot
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
Signed-off-by: Michel Aractingi <michel.aractingi@huggingface.co >
* set num_workers to zero in example
* style nit
* reintroduce copilot optim
---------
Signed-off-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-10-23 17:34:03 +02:00
df71f3ce24
docs(policies): GR00T updates ( #2293 )
...
* Update Libero beval results + fix phrasing
* style of GR00T wording
2025-10-23 15:01:41 +02:00
326aca0a48
Add API Examples ( #2289 )
...
* (unscrewing things up) (#2288 )
* fix: expose a function explicitly building a frame for inference
* fix: first make dataset frame, then make ready for inference
* fix: reducing reliance on lerobot record for policy's ouptuts too
* fix: encapsulating squeezing out + device handling from predict action
* fix: remove duplicated call to build_inference_frame and add a function to only perform data type handling (whole conversion is: keys matching + data type conversion)
* refactor(envs): add custom-observation-size (#2167 )
* fix: add MockMotorBus to MockRobot
* rl: first drafts
* add: all components of HIL SERL
* fix: actor block works
* fix: less friction, less friction
* add: hil-serl complete example
* fix: dataset names
* fix: restructuring example folder
* fix: act works but found bug in how ACT works
* fix: same path for both pre and postprocessors
* fix: paths
* add: example usage for act
* add: using ACT example
* fix: training examples
* fix: using examples
* fix: camera index
* fix: rename workflows into tutorial so that the path of the files is lerobot/examples/tutorial/...
* fix: upload everything in one repo
* fix: model name
* fix: simplify model path
* add: VLAs example
---------
Signed-off-by: Francesco Capuano <74058581+fracapuano@users.noreply.github.com >
* fix: minor fix using named attributes
* fix: change model to act
* fix: named attributes for inference frame building
* fix: minor fixes to smolvla
* fix: small changes to pi0
* remove: old file that should have never been committed (ups sorry sorry)
---------
Signed-off-by: Francesco Capuano <74058581+fracapuano@users.noreply.github.com >
2025-10-23 14:18:13 +02:00
be46bdea8f
feat(policies): add Nvidia Gr00t N1.5 model ( #2292 )
...
* feat(policies): add Nvidia Gr00t N1.5 model
Co-authored-by: lbenhorin <lbenhorin@nvidia.com >
Co-authored-by: Aravindh <aravindhs@nvidia.com >
Co-authored-by: nv-sachdevkartik <ksachdev@nvidia.com >
Co-authored-by: youliangt <youliangt@nvidia.com >
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Pepijn <138571049+pkooij@users.noreply.github.com >
Co-authored-by: Jade Choghari <chogharijade@gmail.com >
* fix(docs): add groot to index
Co-authored-by: sachdevkartik <sachdev.kartik25@gmail.com >
---------
Co-authored-by: lbenhorin <lbenhorin@nvidia.com >
Co-authored-by: Aravindh <aravindhs@nvidia.com >
Co-authored-by: nv-sachdevkartik <ksachdev@nvidia.com >
Co-authored-by: youliangt <youliangt@nvidia.com >
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co >
Co-authored-by: Pepijn <138571049+pkooij@users.noreply.github.com >
Co-authored-by: Jade Choghari <chogharijade@gmail.com >
Co-authored-by: sachdevkartik <sachdev.kartik25@gmail.com >
2025-10-23 13:50:30 +02:00
306429a85b
fix(cameras): opencv camera index casting ( #2286 )
2025-10-22 17:27:31 +02:00
12f2f35760
- Introduce _current_file_start_frame for better tracking of the number of frames in each parquet file ( #2280 )
...
- Added testing for that section in `test_datasets.py`
2025-10-21 16:17:12 +02:00
a024d33750
fix(bug): Fix policy renaming ValueError during training ( #2278 )
...
* fixes
* style
* Update src/lerobot/policies/factory.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
* style
* add review fixes
---------
Signed-off-by: Jade Choghari <chogharijade@gmail.com >
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2025-10-21 16:00:46 +02:00
63cd2111ad
[Fix] Device Error on SmolVLA Multi-GPU Training ( #2270 )
...
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org >
2025-10-21 14:26:31 +02:00
abe9e79825
chore(dependencies): bump & ceil gymnasium version + pin metaworld version + bump gym-hil ( #2267 )
...
* chore(dependencies): bump & ceil gymnasium version + pin metaworld version
Co-authored-by: Jade Choghari <chogharijade@gmail.com >
* chore(dependencies): bump gym-hil to be compatible
---------
Co-authored-by: Jade Choghari <chogharijade@gmail.com >
2025-10-21 12:56:32 +02:00
503fc4e9f4
fix(ci): exclude motor tests in multi-gpu setup ( #2276 )
2025-10-21 12:14:26 +02:00
92b479f9ac
Fix camera FPS set issue ( #2275 )
...
Set camera width/height 1st before FPS setting, to avoid FPS set failure alike:
ERROR:__main__:Failed to connect or configure OpenCV camera /dev/video2: OpenCVCamera(/dev/video2) failed to set fps=30 (actual_fps=25.0).
2025-10-21 11:31:03 +02:00
jade.choghari@huggingface.co