# UAIDE Backend

FastAPI inference service for the user-provided EfficientNet-B2 + FFT fusion image model.

## Run

```bash
python3 -m venv backend/.venv
source backend/.venv/bin/activate
pip install -r backend/requirements.txt
uvicorn backend.model_service:app --host 0.0.0.0 --port 8000
```

## Notes

- Image inference is wired to `integration_assets/best_model_weights.pt`.
- Video uploads currently return a clear `501` until video weights/code are provided.