Basswala/Algerian-Forest-Fires-Regression-Project

## Algerian Forest Fires – Regression Project This repository contains data cleaning, exploratory analysis, and regression modeling notebooks for predicting the Fire Weather Index (FWI) using the Algerian Forest Fires dataset. ### Project Structure ``` Regression_Projects/ ├── data/ │ ├── raw/ # Original datasets (read-only) │ │ └── Algerian_forest_fires_dataset_UPDATE.csv │ └── processed/ # Cleaned/derived datasets │ └── Algerian_forest_fires_cleaned_dataset.csv ├── models/ # Trained model artifacts (optional) ├── notebooks/ # Analysis & modeling notebooks │ ├── Ridge, Lasso Regression.ipynb │ └── Model Training.ipynb └── reports/ └── figures/ # Generated plots/figures ``` ### Environment Setup 1. Python 3.10+ recommended. Create a virtual environment (venv or conda): ```bash python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate pip install -U pip ``` 2. Install dependencies: ```bash pip install pandas numpy matplotlib seaborn scikit-learn jupyter ``` 3. Launch Jupyter: ```bash jupyter notebook ``` Open notebooks from `notebooks/`. ### Data - Raw: `data/raw/Algerian_forest_fires_dataset_UPDATE.csv` - Processed: `data/processed/Algerian_forest_fires_cleaned_dataset.csv` (created by the cleaning steps in notebooks) The notebooks have been updated to use relative paths from the `notebooks/` directory: - Read raw: `../data/raw/Algerian_forest_fires_dataset_UPDATE.csv` - Write processed: `../data/processed/Algerian_forest_fires_cleaned_dataset.csv` - Read processed: `../data/processed/Algerian_forest_fires_cleaned_dataset.csv` ### Notebooks - `Ridge, Lasso Regression.ipynb` - Cleans the dataset, fixes column types, encodes `Classes` labels, performs EDA (histograms, pie chart, correlation heatmap), and saves a processed CSV. - Computes correlations with `numeric_only=True` to avoid errors from st …