README.md

metadata

title: iris-backend
emoji: 🌸
colorFrom: green
colorTo: blue
sdk: docker
pinned: false
license: mit

🌸 Iris Flower Classifier (Flask API)

A lightweight Flask API hosted on Hugging Face Spaces that predicts the species of an Iris flower using a Scikit-learn Logistic Regression model.

The frontend UI is hosted separately on GitHub Pages:
👉 Try it here

---

🚀 How it works

1. User enters flower measurements (sepal & petal length/width) on the frontend.
2. Frontend sends data as JSON to this backend API (/predict endpoint).
3. The Flask app loads a pickled Scikit-learn model and predicts the species.
4. Response is returned as JSON and displayed on the frontend.

---

🔧 Tech Stack

• Python 3.9+
• Flask – Web framework
• Flask-CORS – Allow frontend/backend communication
• Scikit-learn – ML model (Logistic Regression)
• Pickle – Model persistence
• Hugging Face Spaces (Docker SDK) – Deployment

---

📂 File Structure

├── app.py            # Flask API
├── model.pkl         # Pre-trained ML model
├── requirements.txt  # Python dependencies
├── Dockerfile        # Custom Space runtime
└── README.md         # Project documentation

---

📡 API Usage

Endpoint

POST /predict

Request (JSON)

{
  "sepal_length": 5.1,
  "sepal_width": 3.5,
  "petal_length": 1.4,
  "petal_width": 0.2
}

Response (JSON)

{
  "prediction": "setosa"
}

---

🛠️ Run Locally

Clone and run with Docker:

git clone https://huggingface.co/spaces/<your-username>/iris-backend
cd iris-backend
docker build -t iris-backend .
docker run -p 7860:7860 iris-backend

Now open http://localhost:7860/predict.

---

🌐 Frontend

The frontend is hosted on GitHub Pages: 👉 Iris Classifier Frontend

Frontend calls the backend API hosted here on Hugging Face Spaces.

---

📸 Demo Screenshot

---

📜 License

MIT License – free to use and modify.

---