adityapandey2608/FWI-Prediction

A Ml project Based on Algerian Forest Fire Dataset that Predicts FWI # 🔥 Fire Weather Index (FWI) Prediction App This is a Flask web application that predicts the **Fire Weather Index (FWI)** using a machine learning model trained on the **Algerian Forest Fire dataset**. --- ## 📊 Dataset Info - **Source**: Algerian Forest Fire Dataset - **Attributes used**: - Temperature - Relative Humidity (RH) - Wind Speed (Ws) - Rain - FFMC (Fine Fuel Moisture Code) - DMC (Duff Moisture Code) - ISI (Initial Spread Index) - Class (Fire occurrence class) The dataset includes meteorological and fire data from two regions of Algeria collected between June and September 2012. --- ## 🧠 Model Training The model was trained using **Ridge Regression** in scikit-learn. The steps included: 1. Cleaning the dataset (removing nulls, standardizing column names) 2. Encoding categorical variables if needed 3. Scaling features using `StandardScaler` 4. Training a Ridge Regression model on the scaled data 5. Saving the model (`regressor.pkl`) and scaler (`scaler.pkl`) using `pickle` > The training process and data exploration are documented in the `Model Training.ipynb` and `Fire Forest EDA.ipynb` notebooks. --- ## 🚀 Features - Simple web interface for FWI prediction - Scikit-learn Ridge Regression model - Input form for environmental variables - Real-time prediction result - Scaled inputs using `StandardScaler` - Clean user interface with HTML/CSS --- ## 🛠 How to Run Locally ### 1. Clone the repository ```bash git clone github.com cd fwi-predictor