Roland Tannous 48a5e49313 fix: remove Pass 3 self-scoring, trust Pass 2 output directly ... The LLM was bad at scoring its own conversion quality — rejecting good Pass 2 output (score 5/10 for a perfectly usable conversion). Instead: - Remove Pass 3 entirely (saves ~0.4s and one inference call) - Trust Pass 2 output and return it to the user - Build notification from Pass 1 classification info instead - User can always adjust mapping via dropdowns if they disagree		2026-03-10 15:56:48 +00:00
..
__init__.py	feat: add ShareGPT+image VLM format support and improve image column detection	2026-03-10 01:27:36 +00:00
chat_templates.py	Add AGPL-3.0 SPDX headers to all source files	2026-03-09 20:17:45 +00:00
data_collators.py	Add AGPL-3.0 SPDX headers to all source files	2026-03-09 20:17:45 +00:00
dataset_utils.py	fix: harden template mapping for complex column types and curly braces	2026-03-10 15:43:35 +00:00
format_conversion.py	feat: add LLM-assisted dataset detection using ephemeral GGUF helper	2026-03-10 09:20:45 +00:00
format_detection.py	fix: probe image column candidates when multiple exist	2026-03-10 01:38:33 +00:00
llm_assist.py	fix: remove Pass 3 self-scoring, trust Pass 2 output directly	2026-03-10 15:56:48 +00:00
model_mappings.py	Add AGPL-3.0 SPDX headers to all source files	2026-03-09 20:17:45 +00:00
vlm_processing.py	debug: switch to print() for subprocess visibility	2026-03-10 12:49:01 +00:00