README.md

metadata

title: BioClip Image Classification
emoji: 🌿
colorFrom: green
colorTo: blue
sdk: gradio
sdk_version: 3.27.0
app_file: app.py
pinned: false

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

BioClip Image Classification

This Hugging Face Space demonstrates image classification using the BioClip model. Upload an image to get a prediction of its class, along with the top 3 most similar classes and file paths.

How to Use

1. Open the Gradio interface in this Space.
2. Upload an image using the provided input area.
3. The model will process the image and return:
   • The predicted class
   • The top 3 most similar classes
   • The top 3 most similar file paths from the dataset

About the Model

This Space uses the BioClip model, which is designed for biological image classification. The model is loaded from the Hugging Face model hub (imageomics/bioclip).

Technical Details

• The Space uses Gradio for the user interface.
• It employs FAISS indexes for efficient similarity search.
• The classification is performed using a k-nearest neighbors approach with majority voting.

Note

The FAISS indexes are downloaded at runtime. Make sure your Space has internet access to download these files when the app starts.