Shounak-Chavan/Algerian-Forest-Fire-Prediction-ML

Practice ML project using Ridge, Lasso, and ElasticNet models on the Algerian Forest Fire dataset. Includes EDA, preprocessing, feature selection, scaling, and Flask deployment. # 🔥 Algerian Forest Fire — FWI Prediction System ### *(Fire Weather Index Prediction using Ridge & Lasso Regression)* This project is a **Machine Learning–based wildfire risk prediction system** that predicts the **Fire Weather Index (FWI)** — a critical metric for assessing wildfire danger. The system processes meteorological and forest data, applies advanced regression models, and delivers predictions through an interactive **Flask** web application. The workflow includes comprehensive data cleaning, exploratory data analysis (EDA), intelligent feature engineering, correlation-based feature selection, data scaling, multiple regression model implementations, and a professional web UI for real-time predictions. --- ## 🚀 Features - 📊 **Cleaned & Preprocessed** Algerian Forest Fires dataset - 🔥 **FWI Prediction** using meteorological and forest parameters - 📉 **Multiple Models** — Ridge, Lasso, ElasticNet, and Linear Regression - ⚙️ **Production-Ready** — StandardScaler + Model persistence with Pickle - 🌐 **Interactive Flask UI** with Bootstrap styling - 📂 **Clean & Modular** project structure - 📈 **Hyperparameter Tuning** with RidgeCV, LassoCV, ElasticNetCV --- ## 🛠 Tech Stack - **Python** - **Flask** - **NumPy** - **Pandas** - **Scikit-Learn** - **Seaborn** - **Matplotlib** - **Pickle** --- ## 📁 Project Structure ``` Algerian-Forest-Fire-Prediction-ML/ │ ├── app.py ├── requirements.txt ├── README.md ├── .gitignore │ ├── Models/ │ ├── ridge.pkl │ └── scalar.pkl │ ├── templates/ │ └── index.html │ └── Notebooks/ ├── Algerian_forest_fires_dataset_UPDATED.ipynb ├── Cleaned_Dataset ├── EDA.ipynb └── ModelTraining.ipynb ``` --- ## ▶️ How to Run Locally ### **1️⃣ Clone the Repository** ``` git clone github.com cd Algerian-Forest-Fire-Prediction-ML ``` ### **2️⃣ (Optional) Create Virtual Environment** ``` python -m venv venv ``` Activate the venv: **Windows:** ``` venv\Scripts\activate ``` ** …