cmoyates 9cd800d37b fix: reverse decoder concat order to match Torch (c4,c3,c2,c1) ... Torch decoder concatenates feature projections as [c4,c3,c2,c1] but MLX was using [c1,c2,c3,c4]. The linear_fuse conv was trained with the Torch order, producing scrambled features. Verified fix gives correlation=1.0 against Torch on identical inputs. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>		2026-03-01 09:44:41 -03:30
..
convert	feat(phase3): PyTorch→MLX weight converter (#1)	2026-03-01 05:16:56 -03:30
inference	fix: guard against overlap >= tile_size in tiled inference	2026-03-01 06:27:31 -03:30
io	fix: resize inputs to model img_size, use infer_and_save in CLI	2026-03-01 06:11:57 -03:30
model	fix: reverse decoder concat order to match Torch (c4,c3,c2,c1)	2026-03-01 09:44:41 -03:30
utils	feat(phase6): optimization, benchmarking, and tiled inference	2026-03-01 06:24:10 -03:30
__init__.py	feat: add CorridorKeyMLXEngine integration surface	2026-03-01 06:43:07 -03:30
engine.py	fix: remove double normalize_rgb in engine preprocessing	2026-03-01 06:47:22 -03:30