README.md

---
title: UFO Web
emoji: 🛸
colorFrom: blue
colorTo: green
sdk: streamlit
sdk_version: 1.58.0
app_file: app.py
pinned: false
---

# UFO Country Predictor

Session 406 machine-learning web app practice project. It predicts the likely reporting country for a UFO sighting from duration, latitude, and longitude.

## Project Structure

```text
UFO_local/
├── app.py                 # Hugging Face Spaces Streamlit app
├── backend.py             # FastAPI backend for local deployment practice
├── frontend.py            # Streamlit frontend that calls the FastAPI backend
├── requirements.txt
├── ufo-model.pkl
└── course_materials/      # Session 406 task guide and original Flask reference
```

## Local FastAPI + Streamlit Workflow

Install dependencies:

```bash
conda activate session406
pip install -r requirements.txt
```

Start the FastAPI backend:

```bash
uvicorn backend:app --reload --host 127.0.0.1 --port 8001
```

Start the Streamlit frontend in another terminal:

```bash
streamlit run frontend.py --server.address 127.0.0.1 --server.port 8501
```

Open:

```text
http://127.0.0.1:8501
```

## Hugging Face Spaces Workflow

Hugging Face Spaces runs `app.py` directly as a single Streamlit app. This version loads `ufo-model.pkl` itself, so it does not need the FastAPI backend.

Create or open the Space:

```text
https://huggingface.co/spaces/Coco-Spot/UFO-web
```

Push this project to the Space repository:

```bash
git remote add hf https://huggingface.co/spaces/Coco-Spot/UFO-web
git push hf main
```

When prompted, use your Hugging Face username and a write token from:

```text
https://huggingface.co/settings/tokens
```

## Course Materials

The original session files are kept in `course_materials/session-406-ml-app/`, including the UFO Flask practice guide, original solution, data file, and task roadmap.