A Machine Learning project that predicts the Fire Weather Index (FWI) — a numerical indicator of forest fire risk — using the Algerian Forest Fire dataset.
This project demonstrates the end-to-end ML pipeline:
- 📊 Data preprocessing & model building
- ⚙️ API creation with Flask
- ☁️ Deployment on Render with a live frontend
🌐 Live Demo → FWI Predictor Web App
Forest fires pose serious ecological and economic threats. The FWI provides an estimate of fire danger based on weather conditions. This project builds a regression model to predict FWI values and makes it accessible via a web app.
✨ Key Features:
- 🤖 Regression model (Linear Regression as baseline)
- 🛠️ Flask API serving live predictions
- 🌍 Deployed on Render with a simple frontend
- 🔄 Full-stack implementation: ML → API → Deployment
- 🐍 Python (Pandas, NumPy, Scikit-learn)
- 🚀 Flask (for API development)
- 🎨 HTML/CSS/JS (for frontend UI)
- ☁️ Render (for deployment)
├── data/ # Dataset (Algerian Forest Fire dataset)
├── notebooks/ # Jupyter notebooks for EDA & model training
├── model/ # Trained model files
├── app/
│ ├── app.py # Flask API entry point
│ ├── requirements.txt # Dependencies
│ └── templates/ # Frontend (HTML/CSS)
├── README.md # Project documentation
└── deployment/ # Render configuration files
git clone https://github.com/your-username/fwi-predictor.git
cd fwi-predictorpip install -r requirements.txtpython app/app.pyOpen 👉 http://localhost:5000 in your browser.
- 🔎 Improve accuracy with advanced models (Random Forest, Gradient Boosting, Neural Networks).
- 🌦️ Integrate real-time weather APIs for live forecasting.
- ⚡ Explore MLOps practices for automated training and CI/CD pipelines.
This project is open-source and available under the MIT License.
Yash 💼 LinkedIn: Your Profile