wangzixiang
/
lerobot
mirror of https://github.com/huggingface/lerobot.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
892 Commits 123 Branches 0 Tags 259 MiB
Commit Graph

892 Commits

Author SHA1 Message Date
Michel Aractingi 764925e4a2 added temporary fix for missing task_index key in online environment 2025-03-28 17:18:24 +00:00
Michel Aractingi 7bb142b707 split encoder for critic and actor 2025-03-28 17:18:24 +00:00
Michel Aractingi 2c2ed084cc style fixes 2025-03-28 17:18:24 +00:00
KeWang1017 91fefdecfa Refactor SAC configuration and policy for improved action sampling and stability 
- Updated SACConfig to replace standard deviation parameterization with log_std_min and log_std_max for better control over action distributions.
- Modified SACPolicy to streamline action selection and log probability calculations, enhancing stochastic behavior.
- Removed deprecated TanhMultivariateNormalDiag class to simplify the codebase and improve maintainability.

These changes aim to enhance the robustness and performance of the SAC implementation during training and inference.
 2025-03-28 17:18:24 +00:00
KeWang1017 70e3b9248c Refine SAC configuration and policy for enhanced performance 
- Updated standard deviation parameterization in SACConfig to 'softplus' with defined min and max values for improved stability.
- Modified action sampling in SACPolicy to use reparameterized sampling, ensuring better gradient flow and log probability calculations.
- Cleaned up log probability calculations in TanhMultivariateNormalDiag for clarity and efficiency.
- Increased evaluation frequency in YAML configuration to 50000 for more efficient training cycles.

These changes aim to enhance the robustness and performance of the SAC implementation during training and inference.
 2025-03-28 17:18:24 +00:00
KeWang1017 0ecf40d396 Refactor SACPolicy for improved action sampling and standard deviation handling 
- Updated action selection to use distribution sampling and log probabilities for better stochastic behavior.
- Enhanced standard deviation clamping to prevent extreme values, ensuring stability in policy outputs.
- Cleaned up code by removing unnecessary comments and improving readability.

These changes aim to refine the SAC implementation, enhancing its robustness and performance during training and inference.
 2025-03-28 17:18:24 +00:00
KeWang1017 a113daa81e trying to get sac running 2025-03-28 17:18:24 +00:00
Michel Aractingi 80b86e9bc3 Added normalization schemes and style checks 2025-03-28 17:18:24 +00:00
Michel Aractingi 9dafad15e6 added optimizer and sac to factory.py 2025-03-28 17:18:24 +00:00
Eugene Mironov d96edbf5ac [HIL-SERL PORT] Fix linter issues (#588) 2025-03-28 17:18:24 +00:00
Eugene Mironov 6340d9d17c [Port Hil-SERL] Add unit tests for the reward classifier & fix imports & check script (#578) 2025-03-28 17:18:24 +00:00
Michel Aractingi 66268fcf85 added comments from kewang 2025-03-28 17:18:24 +00:00
KeWang1017 a5228a0dfe Enhance SAC configuration and policy with new parameters and subsampling logic 
- Added `num_subsample_critics`, `critic_target_update_weight`, and `utd_ratio` to SACConfig.
- Implemented target entropy calculation in SACPolicy if not provided.
- Introduced subsampling of critics to prevent overfitting during updates.
- Updated temperature loss calculation to use the new target entropy.
- Added comments for future UTD update implementation.

These changes improve the flexibility and performance of the SAC implementation.
 2025-03-28 17:18:24 +00:00
KeWang dbadaae28b Port SAC WIP (#581) 
Co-authored-by: KeWang1017 <ke.wang@helloleap.ai>
 2025-03-28 17:18:24 +00:00
Michel Aractingi 44536d1f0c completed losses 2025-03-28 17:18:24 +00:00
Michel Aractingi 69b6de46d7 nit in control_robot.py 2025-03-28 17:18:24 +00:00
Michel Aractingi 399f834788 Update lerobot/scripts/train_hilserl_classifier.py 
Co-authored-by: Yoel <yoel.chornton@gmail.com>
 2025-03-28 17:18:24 +00:00
Eugene Mironov df57d372d6 Fixup 2025-03-28 17:18:24 +00:00
Michel Aractingi 76234b7d14 Add human intervention mechanism and eval_robot script to evaluate policy on the robot (#541) 
Co-authored-by: Yoel <yoel.chornton@gmail.com>
 2025-03-28 17:18:24 +00:00
Yoel 58cc445921 Reward classifier and training (#528) 
Co-authored-by: Daniel Ritchie <daniel@brainwavecollective.ai>
Co-authored-by: resolver101757 <kelster101757@hotmail.com>
Co-authored-by: Jannik Grothusen <56967823+J4nn1K@users.noreply.github.com>
Co-authored-by: Remi <re.cadene@gmail.com>
Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co>
 2025-03-28 17:18:24 +00:00
Steven Palma b568de35ad
fix(datasets): cast imgs_dir as Path (#915) 2025-03-28 18:08:12 +01:00
Yongjin Cho ae9c81ac39
fix(docs): correct spelling of 'ffmpeg' in installation instructions (#914) 2025-03-28 17:43:33 +01:00
Steven Palma 78fd1a1e04
chore(docs): update docs (#911) 2025-03-27 09:55:06 +01:00
Steven Palma 90533e6b9f
fix(docs): hot-fix updating installation instructions after #883 (#907) 2025-03-26 13:21:40 +01:00
AlexC 2c22f7d76d
Add offline mode in the configuration for wandb logging (#897) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
 2025-03-25 13:44:49 +01:00
Qizhi Chen a774af2eab
fix pi0 action padding name (#893) 
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
 2025-03-25 11:24:46 +01:00
Steven Palma 725b446ad6
fix(deps): constrain PyAV version to resolve OpenCV-python ffmpeg version conflict (#883) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-24 23:40:22 +01:00
Steven Palma a6015a55f9
chore(scripts): remove deprecated script (#887) 2025-03-23 01:16:50 +01:00
Cole f39652707c
add docs details for resolving firmware update issues (#627) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-19 19:17:07 +01:00
Steven Palma 712d5dae4f
fix(os): fix default codec for windows (#875) 2025-03-18 22:04:21 +01:00
Pepijn 952e892fe5
Use float32 instead of int (#877) 2025-03-18 16:36:37 +01:00
Pepijn e8159997c7
User/pepijn/2025 03 17 act different image shapes (#870) 2025-03-18 11:09:05 +01:00
Steven Palma 1c15bab70f
fix(codec): hot-fix for default codec in linux arm platforms (#868) 2025-03-17 13:23:11 +01:00
Guillaume LEGENDRE 9f0a8a49d0
Update test-docker-build.yml 2025-03-15 11:34:17 +01:00
Huan Liu a3cd18eda9
added wandb.run_id to allow resuming without wandb log; updated log m… (#841) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-15 09:40:39 +01:00
Huan Liu 7dc9ffe4c9
Update 10_use_so100.md (#840) 2025-03-14 17:07:14 +01:00
Jade Choghari 0e98c6ee96
Add torchcodec cpu (#798) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Remi <re.cadene@gmail.com>
Co-authored-by: Remi <remi.cadene@huggingface.co>
Co-authored-by: Simon Alibert <simon.alibert@huggingface.co>
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-14 16:53:42 +01:00
Simon Alibert 974028bd28
Organize test folders (#856) 
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
 2025-03-13 14:05:55 +01:00
Simon Alibert a36ed39487
Improve pre-commit config (#857) 2025-03-13 13:29:55 +01:00
Ermano Arruda c37b1d45b6
parametrise tolerance_s in visualize_dataset scripts (#716) 2025-03-13 10:28:29 +01:00
pre-commit-ci[bot] f994febca4
[pre-commit.ci] pre-commit autoupdate (#844) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-03-11 11:28:01 +01:00
Steven Palma 12f52632ed
chore(docs): update instructions for change in device and use_amp (#843) 2025-03-10 21:03:33 +01:00
Steven Palma 8a64d8268b
chore(deps): remove hydra dependency (#842) 2025-03-10 19:00:23 +01:00
Pepijn 84565c7c2e
Fix camera rotation error (#839) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-03-10 17:02:19 +01:00
Ben Sprenger 05b54733da
feat: add support for external plugin config dataclasses (#807) 
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
 2025-03-10 13:25:47 +01:00
Simon Alibert 513b008bcc
fix: deactivate tdmpc backward compatibility test with use_mpc=True (#838) 2025-03-10 10:19:54 +01:00
Joe Clinton 32fffd4bbb
Fix delay in teleoperation start time (#676) 
Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2025-03-08 11:40:07 +01:00
Simon Alibert 03c7cf8a63
Remove pr_style_bot (#832) 2025-03-08 09:39:07 +01:00
Simon Alibert 074f0ac8fe
Fix gpu nightly (#829) 2025-03-07 13:21:58 +01:00
Mathias Wulfman 25c63ccf63
🐛 Remove `map_location=device` that no longer exists when loading DiffusionPolicy from_pretained after commit 5e94738 (#830) 
Co-authored-by: Mathias Wulfman <mathias.wulfman@wandercraft.eu>
 2025-03-07 13:21:11 +01:00