Suman991/Fire-predictor

This project predicts the Forest Weather Index (FWI) for Algerian forest fire datasets using a trained Ridge Regression model. It includes data preprocessing, model training, and a Flask-based web application for deployment. Users can input parameters like Temperature, RH, Wind Speed, Rain, and more to get real-time fire index predictions. # Algerian Forest Fire Predictor ## Project Overview This project predicts the **Forest Weather Index (FWI)** for Algerian forest fire data. The FWI is a numeric value that indicates the severity and spread potential of forest fires based on environmental parameters such as temperature, humidity, wind speed, and more. The model achieves an **R² score of 0.9842** after extensive experimentation with various regression techniques, providing highly accurate predictions. --- ## Features - **Input Parameters:** - **Temperature**: Air temperature in Celsius. - **RH**: Relative Humidity in percentage. - **Ws**: Wind speed in km/h. - **Rain**: Rainfall in mm. - **FFMC**: Fine Fuel Moisture Code. - **DMC**: Duff Moisture Code. - **ISI**: Initial Spread Index. - **Classes**: Fire severity class. - **Region**: Location-based forest fire region. - **Output**: Predicted **FWI** (Forest Weather Index). - **Model Used**: Linear Regression ( Ridge , Lasso, ElasticNet). - **Data Preprocessing**: StandardScaler is used to scale input features for model training and prediction. --- ## Model Performance - **R² Score**: 0.9842 - **Techniques Tried**: Various linear regression models, including Ridge , Lasso , ElasticNet regression, were applied to optimize prediction accuracy. --- ## Tech Stack - **Backend**: Python (Flask Framework) - **Machine Learning**: Scikit-Learn - **Frontend**: HTML, CSS - **Model Storage**: Pickle file (.pkl) --- ## Installation Guide Follow these steps to set up the project: ### 1. Clone the Repository ```bash git clone github.com cd Fire-predictor ``` ### 2. Install Dependencies Make sure Python is installed. Install required libraries: ```bash pip install -r requirements.txt ``` *Requirements include Flask, scikit-learn, and numpy.* ### 3. Run the Flask Application Start the application server: ```bash python app.py ``` The application will run on `127.0.0.1`. --- ## File Structure ``` project-root/ …