Skip to content

Latest commit

 

History

History
59 lines (46 loc) · 2.58 KB

README.md

File metadata and controls

59 lines (46 loc) · 2.58 KB

Project Overview

This repository combines two related projects focusing on housing price predictions using the Ames Housing Dataset. The first project emphasizes innovative feature engineering to create new variables for enhanced predictive power. The second project applies a Stochastic Gradient Descent Regression (SGD) model with rigorous hyperparameter tuning to evaluate model performance and provide actionable insights.

Objectives:

  1. Feature Engineering:

    • Engineer three unique features that add predictive value to housing price models.
    • Explore the correlative impact of these features on the response variable (Sale Price).

  2. Model Tuning:

    • Develop a Stochastic Gradient Descent Regression model.
    • Conduct hyperparameter tuning to optimize predictive accuracy.
    • Evaluate and interpret model metrics, including R-squared and residual error.

Technologies Used

  • Python:
    • Libraries: pandas, numpy, scikit-learn, matplotlib, seaborn
  • Jupyter Notebook: Analysis and modeling workflows
  • Excel: Dataset preparation and visualization

Repository Structure

  • /data:
    • AmesHousing.csv: The original Ames Housing Dataset.
  • /notebooks:
    • Feature_Engineering_Ames.ipynb: Notebook detailing the feature engineering process.
    • Model_Discovery_and_Tuning.ipynb: Notebook with model tuning and evaluation steps.
  • /documentation:
    • Model_Discovery_and_Tuning.html: Html File with model tuning and evaluation steps.

Key Insights

  1. Feature Engineering:

    • Engineered features (e.g., Lot Frontage Ratio, Age of Home, and Neighborhood Premium Index) showed significant positive correlations with housing prices.
    • Practical real-world implications, such as identifying neighborhoods with higher premiums or properties with favorable characteristics.

  2. Model Performance:

    • The SGD model achieved strong predictive performance with optimal hyperparameters.
    • Hyperparameter tuning minimized overfitting and balanced R-squared values between training and testing sets.

  3. Actionable Insights:

    • High-performing neighborhoods can be targeted for investment or development.
    • Importance of engineered features in capturing nuanced patterns in real estate data.

Instructions

Requirements

  • Python 3.8+
  • Libraries: pandas, numpy, scikit-learn, matplotlib, seaborn

Running the Project

  1. Clone the repository: 
    git clone https://github.com/stefagnone/Ames-Housing-Feature-Engineering-and-Model-Tuning.git
cd Ames-Housing-Feature-Engineering-and-Model-Tuning