Fuel Economy Prediction Analysis

This project predicts vehicle fuel efficiency (MPG) using the classic Auto-MPG dataset.

Key Results

• Final Test R2: 0.832
• Mean Absolute Error (MAE): 2.52 MPG

Technical Workflow

1. EDA: Identified multicollinearity between engine displacement and weight.
2. Feature Engineering: Created Polynomial Features and Interaction Terms.
3. Transformation: Applied Log Transformation to the target variable to address heteroscedasticity.

How to Run

1. Install dependencies: pip install -r requirements.txt
2. Run the Jupyter Notebook: fuel_prediction.ipynb