Add model card and metadata

by nielsr HF Staff - opened Jan 15

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+27

-1

nielsr

Jan 15

Hi! I'm Niels from the Hugging Face community team. I noticed this repository is currently a placeholder.

This PR adds a proper model card, linking it to the research paper Comprehensive language-image pre-training for 3D medical image understanding, and includes the relevant pipeline_tag metadata to improve discoverability on the Hub.

Add model card and metadatacc8e9020

fepegar

Microsoft org Jan 15

Hi, @nielsr . Thanks for your contribution (not sure whether this PR was automated).

The whole repo is actually ready, with a comprehensive model card etc. I'm just waiting for our legal team to give us the green light, which will hopefully happen within days from now, so I can push everything to the hub.

fepegar

Microsoft org Jan 15

Can you help me understand why we need the image-text-to-text pipeline tag? This is basically a CLIP model for CT scans.

nielsr

Jan 17

Ok, yeah that tag is mainly used for VLMs. For CLIP-like models, we usually use zero-shot-image-classification: https://huggingface.co/models?pipeline_tag=zero-shot-image-classification&sort=trending

fepegar

Microsoft org Jan 19

Noted and edited. I had feature-extraction, but zero-shot-image-classification is indeed more appropriate. Thank you, @nielsr .

fepegar changed pull request status to closed Jan 19

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment