ModelPublicadoExterno

Scalesurfer v8

This repository contains the ScaleSurfer 3D TransUNet volume segmentation checkpoint fine-tuned for FreeSurfer v8 labels.

Fuentehuggingface.co/rphammonds/scalesurfer-v8/tree/5764b690df77604c76db0565eb371b1926cc81b6

TipoOtherVersiónRevision 5764b69Bcalidad

Fuente del modelo

Tipo: Other
Tamaño del archivo: 10,0 MB
Versión: vRevision 5764b69
Actualizado: 1 nov 2026
Descargas: 5
Fuente: Hugging Face

Descripción de la fuente

ScaleSurfer FreeSurfer v8 Volume Model

This repository contains the ScaleSurfer 3D TransUNet volume segmentation checkpoint fine-tuned for FreeSurfer v8 labels.

Files

transunet3d.safetensors: model weights in safetensors format.
config.json: architecture and checkpoint configuration.
metadata.json: conversion checksums and sanitized training/evaluation metadata.

Metrics

Split counts	Best validation CE	Final test CE
train=386, val=48, test=48	0.04181101879415413	0.03888384365321448

Loading

ScaleSurfer resolves this model automatically when ScaleSurfer(..., fs_version=8) is used.

This model is intended for research workflows and is not a clinical diagnostic device.

Fuentes

1 fuente

Hugging FacePrincipalVersión 5764b690df77604c76db0565eb371b1926cc81b6

huggingface.co/rphammonds/scalesurfer-v8/tree/5764b690df77604c76db0565eb371b1926cc81b6

Verificado 1 nov

Download files

Repository files

File inventory only; completeness is unverified.

Artefactos del modelo

1 artefacto

transunet3d.safetensors

safetensors · FP32 (unverified) · 10,0 MB · SHA-256 13d135a87c5d…ed69 · Hugging Face

Download file

Extractos de fuentes

2 extractos

Readme18 agohuggingface.co/rphammonds/scalesurfer-v8

--- library_name: pytorch pipeline_tag: image-segmentation tags: - medical-imaging - freesurfer - segmentation - safetensors - scalesurfer --- # ScaleSurfer FreeSurfer v8 Volume Model This repository contains the ScaleSurfer 3D TransUNet volume segmentation checkpoint fine-tuned for FreeSurfer v8 labels. ## Files - `transunet3d.safetensors`: model weights in safetensors format. - `config.json`: architecture and checkpoint configuration. - `metadata.json`: conversion checksums and sanitized training/evaluation metadata. ## Metrics | Split counts | Best validation CE | Final test CE | | --- | ---: | ---: | | train=386, val=48, test=48 | 0.04181101879415413 | 0.03888384365321448 | ## Loading ScaleSurfer resolves this model automatically when `ScaleSurfer(..., fs_version=8)` is used. This model is intended for research workflows and is not a clinical diagnostic device.

Note18 agohuggingface.co/rphammonds/scalesurfer-v8

Source context: 5 downloads · 0 likes · Pipeline image-segmentation · Library pytorch · Repo rphammonds/scalesurfer-v8