# Minimal Hugging Face Space for Intel Image Classifier

This folder contains a minimal Hugging Face Space app using a custom `index.html` frontend and a Python backend.

## Files
- `app.py` — FastAPI backend for prediction and frontend delivery
- `index.html` — frontend UI for uploading images and selecting the model
- `requirements.txt` — dependencies
- `models/` — place your trained models here

## Model files required
Place your trained models in `hf_space/models/`:

- `pytorch_model.pth`
- `model_best.keras`

## Deploying on Hugging Face Spaces
1. Create a new Space on Hugging Face
2. Select `Python` SDK
3. Push the contents of this `hf_space/` directory to the new Space repository

The Space will run `app.py` automatically and serve `index.html` as the frontend.

## Docker support
A `Dockerfile` is included so you can also build and run the app locally in a container.

## Local testing
From `hf_space/`:

```bash
pip install -r requirements.txt
python app.py
```

Then open:

- `http://localhost:7860`