ydy0615/lerobot-clone
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-06-03 12:21:27 +00:00
Code Issues Packages Projects Releases Wiki Activity
851 Commits 207 Branches 9 Tags
7fd8b4c773db9a3f1366e1d5cd702ee344fd08e4
Commit Graph

114 Commits

Author SHA1 Message Date
nemo
98856662c1 Add basic support for PEFT adapter methods 
This changes adds support for training policies with much less parameters
by applying adapter methods such as LoRA on specific parts of the policies
and therefore possibly higher learning rates / batch sizes.

To make this as accessible as possible I thought it useful to provide
defaults for `target_modules` and `modules_to_save`. Currently only SmolVLA
has such defaults but when we agree that this change is useful I will set
out to generate more such defaults. While the user can override these
settings, they are expected to only change the peft_method, rank and init_type
parameters.
 2025-06-22 13:45:07 +02:00
mshukor
1c873df5c0 Support for PI0+FAST (#921) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Dana Aubakirova <118912928+danaaubakirova@users.noreply.github.com>
Co-authored-by: Remi <re.cadene@gmail.com>
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
 2025-04-04 11:51:11 +02:00
Steven Palma
5e9473806c refactor(config): Move device & amp args to PreTrainedConfig (#812) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-06 17:59:28 +01:00
Simon Alibert
a1809ad3de Add typos checks (#770) 2025-02-25 23:51:15 +01:00
Simon Alibert
b86a2c0b47 Fix wandb logging (#730) 2025-02-14 18:00:12 +01:00
Simon Alibert
e71095960f Fixes following #670 (#719) 2025-02-12 12:53:55 +01:00
Simon Alibert
90e099b39f Remove offline training, refactor train.py and logging/checkpointing (#670) 
Co-authored-by: Remi <remi.cadene@huggingface.co>
 2025-02-11 10:36:06 +01:00
Remi
638d411cd3 Add Pi0 (#681) 
Co-authored-by: Simon Alibert <simon.alibert@huggingface.co>
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
Co-authored-by: Pablo <pablo.montalvo.leroux@gmail.com>
 2025-02-04 18:01:04 +01:00
Simon Alibert
3c0a209f9f Simplify configs (#550) 
Co-authored-by: Remi <remi.cadene@huggingface.co>
Co-authored-by: HUANG TZU-CHUN <137322177+tc-huang@users.noreply.github.com>
 2025-01-31 13:57:37 +01:00
Simon Alibert
32eb0cec8f Dataset v2.0 (#461) 
Co-authored-by: Remi <remi.cadene@huggingface.co>
 2024-11-29 19:04:00 +01:00
Remi
77478d50e5 Refactor record with add_frame (#468) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2024-10-16 20:51:35 +02:00
Alexander Soare
9c7649f140 Make sure init_hydra_config does not require any keys (#376) 2024-08-23 12:27:08 +01:00
Zhuoheng Li
a2592a5563 Provide more information to the user (#358) 
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
Co-authored-by: Remi <re.cadene@gmail.com>
 2024-08-23 11:00:35 +01:00
Alexander Soare
f8a6574698 Add online training with TD-MPC as proof of concept (#338) 2024-07-25 11:16:38 +01:00
Alexander Soare
5f5efe7cb9 Improve error message when attempting to overwrite a training output folder (#322) 2024-07-16 16:50:31 +01:00
Seungjae Lee
7d1542cae1 Add VQ-BeT (#166) 2024-06-26 08:55:02 +01:00
Alexander Soare
9aa4cdb976 Checkpoint on final step of training even when it doesn't coincide with save_freq. (#284) 2024-06-20 08:27:01 +01:00
Thomas Wolf
11f1cb5dc9 Bug fix: fix setting different learning rates between backbone and main model in ACT policy (#280) 2024-06-18 13:31:35 +01:00
Wael Karkoub
54c9776bde Improves Type Annotations (#252) 2024-06-10 19:09:48 +01:00
Alexander Soare
21f222fa1d Add out_dir option to eval (#244) 2024-06-04 21:01:53 +02:00
amandip7
33362dbd17 Adding parameter dataloading_s to console logs and wandb for tracking… (#243) 
Co-authored-by: Remi <re.cadene@gmail.com>
 2024-06-04 17:02:05 +01:00
Radek Osmulski
504d2aaf48 add EpisodeAwareSampler (#217) 
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
 2024-05-31 13:43:47 +01:00
Alexander Soare
111cd58f8a Add MultiLerobotDataset for training with multiple LeRobotDatasets (#229) 2024-05-30 16:12:21 +01:00
Remi
265b0ec44d Refactor env to add key word arguments from config yaml (#223) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2024-05-30 13:45:22 +02:00
Alexander Soare
e3b9f1c19b Add resume training (#205) 
Co-authored-by: Remi <re.cadene@gmail.com>
 2024-05-28 12:04:23 +01:00
Alexander Soare
b6c216b590 Add Automatic Mixed Precision option for training and evaluation. (#199) 2024-05-20 18:57:54 +01:00
Alexander Soare
2b270d085b Disable online training (#202) 
Co-authored-by: Remi <re.cadene@gmail.com>
 2024-05-20 18:27:54 +01:00
Alexander Soare
68c1b13406 Make policies compatible with other/multiple image keys (#149) 2024-05-16 13:51:53 +01:00
Simon Alibert
f52f4f2cd2 Add copyrights (#157) 2024-05-15 12:13:09 +02:00
Alexander Soare
1249aee3ac Enable logging all the information returned by the forward methods of policies (#151) 2024-05-10 07:45:32 +01:00
Simon Alibert
f5e76393eb Release cleanup (#132) 
Co-authored-by: Kashif Rasul <kashif.rasul@gmail.com>
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com>
Co-authored-by: Cadene <re.cadene@gmail.com>
 2024-05-06 03:03:14 +02:00
Alexander Soare
f3bba0270d Remove EMA model from Diffusion Policy (#134) 2024-05-05 11:26:12 +01:00
Simon Alibert
c77633c38c Add regression tests (#119) 
- Add `tests/scripts/save_policy_to_safetensor.py` to generate test artifacts
- Add `test_backward_compatibility to test generated outputs from the policies against artifacts
 2024-05-04 16:20:30 +02:00
Alexander Soare
bccee745c3 Refactor eval.py (#127) 2024-05-03 17:33:16 +01:00
Alexander Soare
d1855a202a Refactor TD-MPC (#103) 
Co-authored-by: Cadene <re.cadene@gmail.com>
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2024-05-01 16:40:04 +01:00
Alexander Soare
a4891095e4 Use PytorchModelHubMixin to save models as safetensors (#125) 
Co-authored-by: Remi <re.cadene@gmail.com>
 2024-05-01 16:17:18 +01:00
Alexander Soare
9d60dce6f3 Tidy up yaml configs (#121) 2024-04-30 16:08:59 +01:00
Simon Alibert
2765877f28 Speed up CI, add more checks (#115) 
- Split pytest and end-to-end tests into separate jobs
- Add poetry check to ensure pyproject.toml and poetry.lock are in sync
- Add ruff format --diff to ensure style formatting is applied (fails if ruff would reformat anything)
 2024-04-29 23:05:55 +02:00
Quentin Gallouédec
508bd92d03 Remove update method from the policy (#99) 
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
 2024-04-29 12:27:58 +02:00
Simon Alibert
791506dfb8 Remove warnings (#111) 
- Replace `use_pretrained_backbone` with `pretrained_backbone_weights`
- Bump diffusers' minimum version `0.26.3` -> `0.27.2`
- Add ignore flags in CI's pytest
- Change Box observation spaces in simulation environments
- Set `version_base="1.2"` in Hydra initializations
- Bump einops' minimum version `0.7.0` -> `0.8.0`
 2024-04-29 00:31:33 +02:00
Alexander Soare
45f351c618 Make sure targets are normalized too (#106) 2024-04-26 11:18:39 +01:00
Remi
e760e4cd63 Move normalization to policy for act and diffusion (#90) 
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
 2024-04-25 11:47:38 +02:00
Remi
c1bcf857c5 Fix online training (#94) 2024-04-23 18:54:55 +02:00
Remi
1030ea0070 Loads episode_data_index and stats during dataset __init__ (#85) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
 2024-04-23 14:13:25 +02:00
Alexander Soare
e2168163cd Quality of life patches for eval.py (#86) 2024-04-19 12:33:47 +01:00
Simon Alibert
7ad1909641 Tests cleaning & simplification (#81) 2024-04-18 14:47:42 +02:00
Remi
0928afd37d Improve dataset examples (#82) 
Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>
 2024-04-18 11:43:16 +02:00
Alexander Soare
bff4b673c9 Merge remote-tracking branch 'upstream/main' into unify_policy_api 2024-04-17 08:08:57 +01:00
Cadene
36d9e885ef Address comments 2024-04-16 17:20:54 +00:00
Cadene
c7a8218620 typo 2024-04-16 17:19:39 +00:00