mirror of
https://github.com/huggingface/lerobot.git
synced 2026-06-23 19:27:08 +00:00
ea908c0672
Implement the RECAP advantage scoring module as a new phase in lerobot-annotate. Uses a frozen distributional VF to compute per-frame advantages, binarizes into positive/negative indicators with per-task threshold, and writes style=advantage persistent rows for policy conditioning. Skips VF inference on intervention frames as an optimization.