This repository contains a model using the California Housing Dataset. The project is about to create a Linear Regression model with Stochastic Gradient Descent with specific penalty parameters and compare its accuracy with a Linear Regression model.
b. Objectives:
- Read the data.
- Exploratory Data Analysis.
- Feature selection.
- Using K-cross validation with K as 5, building 3 linear regression models and comparing their performance:
- First model: Linear Regression model with Stochastic Gradient Descent with a penalty of 'Elastic Net'.
- Second model: Linear Regression model with Stochastic Gradient Descent with a penalty of 'Ridge Regression'.
- Third model: Ordinaty Linear Regression model.
c. Data: The data we will use is the California Housing Dataset from sklearn datasets, StatLib repository. We can see as follow the description of the variables: