Skip to content

Algorithms for logistic regression, including regularization, soft-max loss and classifier

Notifications You must be signed in to change notification settings

lowhung/regression-regularization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Regression, regularization, classification using least squares, basis, RBF, soft-max

About

This project was done during my time studying in the CS 340: Machine Learning course run by Dr. Mike Gelbart and co. Much of the base code can be attributed to him and his team. Implementations of the listed items were done by Matthew Hounslow.

Contents

In this repo you will find working implementations of the soft-max classifier, least squares with polynomial basis, weighted least squares, least squares with RBF, logistic regression (both with and without L0/L1/L2-regularization), examples of cross-validation are also shown. All code is written in Python 3.6. This repo comes with sample data that the aforementioned techniques can be employed on. Errors and analysis are printed to the console.

Dependencies

  • numpy
  • Sklearn
  • scipy
  • matplotlib

Running the project

In order to run the project, use python3 main.py -q <topic-number> where represents the section in main.py. Each section number pertains to a different technique in this case. More comments will be added to these files in the future to give greater clarity.

Releases

No releases published

Packages

No packages published

Languages