Files
Pepijn 8455efc474 feat(umi): simplify to derive_state_from_action and cam0-only
- Remove fix_dataset.py (user fixes dataset at source)
- evaluate.py: replace observation.pose/joints with observation.state
  (8D, derived from action during training, from FK at inference)
- evaluate.py: remove cam1 — training uses only cam0
- docs: rewrite workflow around derive_state_from_action=true,
  updated recompute-stats and training commands with
  relative_exclude_joints for gripper dims

Made-with: Cursor
2026-04-02 15:02:20 +02:00
..
2026-04-01 13:48:06 +02:00
2025-06-05 17:48:43 +02:00
2026-01-06 18:17:29 +01:00
2025-09-29 10:49:36 +02:00
2026-01-27 15:31:53 +01:00
2025-08-01 17:39:39 +02:00
2025-08-01 17:39:39 +02:00
2025-08-01 17:39:39 +02:00
2025-08-01 17:39:39 +02:00
2025-09-15 09:53:30 +02:00