The program a3.py implements linear regression on a dataset with multiple techniques. Information about the dataset is available in the readme.txt file.

Normal equations method - Penrose pseudoinverse matrix is computed using python libraries

Gradient descent method - Loss function is defined as half the sum of squared errors over the training set. Testing set contains 2000 input points and normalisation is used.

Regularization - L1 and L2 regularization techniques are used. 1500 data points are set aside for cross validation, in order to compute the best value for the regularization coefficient.

Libraries used: numpy, pandas, matplotlib

Image files contain the plots generated for L1 and L2 regularization.