yanggangu

/

CLIP-ViT-base-patch32-SMAT-SVHN

Like 0

This PR updates the model card metadata:

• Adds pipeline_tag: image-classification so the model is discoverable under its intended task.
• Removes arxiv:2609.33437 from the YAML tags, since the arXiv ID should only be referenced in the markdown content and the paper link is already present there.
• Does not add a library_name, because encoder.pt is a vision-encoder state dictionary intended to be loaded with the SMAT code and the matching OpenAI CLIP base model, rather than through a directly supported library snippet.

The markdown content is unchanged.