Add crop_outer_slices: true to Calgary inference configs (match zoo val)

Fix ConjGradNet Hub pack: restore conv_out.bias weights and set unet_conv_out_bias: true

Fix ConjGradNet Hub pack: restore conv_out.bias weights and set unet_conv_out_bias: true

Fix ConjGradNet Hub pack: restore conv_out.bias weights and set unet_conv_out_bias: true

Fix ConjGradNet Hub pack: restore conv_out.bias weights and set unet_conv_out_bias: true

Resave Calgary checkpoints for current ModConv Unet/MWCNN key layout

Upload rim_10x.yaml with huggingface_hub

Upload rim_5x.yaml with huggingface_hub

Upload jointicnet_10x.yaml with huggingface_hub

Upload jointicnet_5x.yaml with huggingface_hub

Docs: install from GitHub via conda + editable .[dev], not PyPI

Add DIRECT JOSS citation to model card

Rewrite model card: clearer docs, dataset links, no checkpoint iters

Docs: training protocol and single-R inference masking note

Replace configs with minimal inference-only YAMLs (no training/validation)

Add Calgary-Campinas zoo checkpoints as name_accel.yaml + name_accel.pt

Fix direct predict usage: output_directory is positional

initial commit