Advanced Multi-Output Regression | California Housing Dataset

Project Overview

This project demonstrates advanced multi-output regression on the California Housing Dataset.
It predicts:

1. Median House Value
2. Value per Median Income (engineered secondary target)

The project is implemented from scratch using *gradient descent, **polynomial & interaction features, and *feature scaling.

---

Key Features

• Multi-output regression predicting two targets simultaneously
• Gradient descent implemented from scratch
• Polynomial and interaction features to capture non-linear relationships
• Feature scaling for faster convergence
• Cost convergence and Actual vs Predicted plots
• Feature importance analysis

---

Skills Demonstrated

• Multiple Linear Regression
• Gradient Descent Optimization
• Feature Engineering
• Polynomial Regression
• Multi-output Regression
• Data Visualization
• Model Interpretability

---

Folder Structure