lerobot/lerobot/common/policies
AdilZouitine 54c48b13fd Refactor SAC policy with performance optimizations and multi-camera support
- Introduced Ensemble and CriticHead classes for more efficient critic network handling
- Added support for multiple camera inputs in observation encoder
- Optimized image encoding by batching image processing
- Updated configuration for ManiSkill environment with reduced image size and action scaling
- Compiled critic networks for improved performance
- Simplified normalization and ensemble handling in critic networks
Co-authored-by: michel-aractingi <michel.aractingi@gmail.com>
2025-03-24 09:22:38 +00:00
..
act Update pre-commits (#733) 2025-02-15 15:51:17 +01:00
diffusion Update pre-commits (#733) 2025-02-15 15:51:17 +01:00
hilserl Changed the init_final value to center the starting mean and std of the policy 2025-03-24 09:22:38 +00:00
pi0 Fixes following #670 (#719) 2025-02-12 12:53:55 +01:00
sac Refactor SAC policy with performance optimizations and multi-camera support 2025-03-24 09:22:38 +00:00
tdmpc Update pre-commits (#733) 2025-02-15 15:51:17 +01:00
vqbet Update pre-commits (#733) 2025-02-15 15:51:17 +01:00
__init__.py Add Pi0 (#681) 2025-02-04 18:01:04 +01:00
factory.py Change SAC policy implementation with configuration and modeling classes 2025-03-24 09:11:14 +00:00
normalize.py - Fixed big issue in the loading of the policy parameters sent by the learner to the actor -- pass only the actor to the `update_policy_parameters` and remove `strict=False` 2025-03-24 09:22:38 +00:00
pretrained.py Remove offline training, refactor `train.py` and logging/checkpointing (#670) 2025-02-11 10:36:06 +01:00
utils.py Simplify configs (#550) 2025-01-31 13:57:37 +01:00