Advanced Linear Regression Project

Welcome to the Advanced linear regression Project!

In this project, you will demonstrate what you have learned in this course by conducting an experiment dealing with House Prices.

We have seen in the lectures what are the shortcomings of linear regression and how Regularization helps in controlling Overfit and give the better fir to data. In this exercise we will stepwise write to functions to implement our own Regularization algorithm.

What we have learnt so far..

• Shortcomings of linear regression
• Polynomial Basis Function
• Regularization(L1/L2)
• Bias-variance trade-off

Dataset

Now, we will try to implement the Rigde and lasso on the house_prices_multivariate dataset. We have been working with this dataset for some time now. We applied linear regression and Polynomial linear regression to predict the SalePrice.

Features:

• Total BsmtSF : Total square feet of basement area
• Lot Area :Lot size in square feet
• Street : Type of road access to property
• OverallQual :Rates the overall material and finish of the house
• GrLivArea :Above grade (ground) living area square feet
• GarageCars :Size of garage in car capacity

What you will learn solving this ?

• Learn to Write Sklearn Algorithm for Ridge and Lasso using Cross Validation
• When to apply Regularization technique, how it works and prevent under or over fitting
• How it helps to balance the bais and variance trade-off.

• Benchmark the performances of linear regression against that of Regularization.
• Learn to use cross validation.

Seems like you are all fired up to put a test to your knowledge.

Let's get started!