Snegamurugan15/algerian-forest-fire-regression

Coursework regression analysis for Algerian Forest Fire FWI prediction. # Algerian Forest Fire FWI Regression Coursework project for analyzing the Algerian Forest Fires dataset and predicting Fire Weather Index (FWI) using regression models. ## Contents - `Assignment 1.ipynb`: notebook with preprocessing, exploratory analysis, model training, evaluation, and saved-model loading examples. - `linear_model.pkl`: saved linear regression model. - `poly_model.pkl`: saved polynomial regression model. - `poly_features.pkl`: saved polynomial feature transformer. - `lasso_model.pkl`: saved tuned Lasso model. - `ridge_model.pkl`: saved tuned Ridge model. ## Dataset The notebook expects a cleaned Algerian Forest Fires CSV. The dataset is not committed to this repository. Place `Algerian_forest_fires_cleaned.csv` locally and update the notebook path if needed. Expected target: - `FWI` Important features used in the notebook include fire-weather measurements such as `Temperature`, `RH`, `Ws`, `Rain`, `FFMC`, `DMC`, `DC`, `ISI`, and `BUI`. ## Setup ```bash python -m venv .venv .venv\Scripts\activate pip install -r requirements.txt ``` On macOS or Linux: ```bash source .venv/bin/activate ``` ## Run ```bash jupyter notebook "Assignment 1.ipynb" ``` Run the notebook cells after placing the dataset locally. The saved `.pkl` files are included for coursework review and model-loading demonstrations. ## Status This repository is kept public as a coursework portfolio artifact. It is not intended as a production model service.