5 Commits

Author	SHA1	Message	Date
Pepijn	9b211a45d6	fix: disable thinking mode in Qwen35VL single-episode fallback path ... The single-episode `segment_skills` method was missing `enable_thinking=False` in `apply_chat_template`, causing the model to output reasoning traces instead of JSON when the batch path fails and falls back to per-episode processing. Made-with: Cursor	2026-03-30 15:31:18 +02:00
root	a6387da464	add license	2026-03-11 23:14:22 +00:00
Jade Choghari	0328b3f4aa	Update src/lerobot/data_processing/data_annotations/vlm_annotations.py ... Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Signed-off-by: Jade Choghari <chogharijade@gmail.com>	2026-03-11 16:10:37 -07:00
root	819c1b9710	add tests/fixes	2026-03-11 22:49:06 +00:00
root	f0848c6887	add subtasl	2026-03-11 19:51:48 +00:00