Arpit-mhjn1/algerian-forest-fire-prediction

# 🔥 Algerian Forest Fire Prediction System 🌲 An end-to-end, **📱 phone-friendly** and responsive Machine Learning web application to predict the occurrence of forest fires in two regions of Algeria (**Bejaia** and **Sidi Bel-abbes**) based on meteorological variables. --- ## 📁 Project Structure ```text algerian-forest-fire-prediction/ │── data/ │ ├── raw/ # 📂 Original dataset │ └── processed/ # 🧹 Cleaned & scaled features │── notebooks/ # 📓 Jupyter notebooks for EDA & prototyping │── models/ # 💾 Pickled models, scalers, and evaluation metrics │── src/ # ⚙️ Source code (preprocessing, training pipelines) │── app.py # 🚀 Streamlit web application dashboard │── requirements.txt # 📦 Python dependencies │── Procfile # ☁️ Deployment config for Render/Heroku │── README.md # 📖 Project documentation ``` --- ## ⚙️ Setup & Local Installation 1️⃣ **Clone the repository**: ```bash git clone cd algerian-forest-fire-prediction ``` 2️⃣ **Install dependencies**: ```bash pip install -r requirements.txt ``` 3️⃣ **Data Processing**: Fetch the dataset and run the automated preprocessing pipeline: ```bash python src/preprocess.py ``` 4️⃣ **Model Training**: Train multiple machine learning algorithms and automatically select and serialize the best performing model: ```bash python src/train.py ``` 5️⃣ **Run the Web Application**: Launch the responsive Streamlit dashboard locally: ```bash streamlit run app.py ``` --- ## ✨ Key Features - 📱 **Phone-friendly & Responsive UI**: Premium *"Dark Fire"* forest aesthetic with custom Streamlit styling, structured card layouts, and complete phone friendliness (optimized touch targets, scaling typography, and responsive margins across phones, tablets, and desktop displays). - 📊 **Interactive Plotly Visualizations**: Features zoomable, interactive, and phone-friendly charts (stacked vertically with horizo …