/GURLS

GURLS: a Least Squares Library for Supervised Learning

Primary LanguageC++

----------------------------------------------------------
-------- Grand Unified Regularized Least Squares ---------
----------------------------------------------------------

Table of Contents
=================

- Introduction
- Documentation

Introduction
============

The GRAND UNIFIED REGULARIZED LEAST SQUARES software library comprises the following packages.

-GURLS, a MATLAB software library for regression and (multiclass) classification 
 based on the Regularized Least Squares (RLS) loss function. 
 Datasets that fit into your computer's memory should be handled with this package.

-bGURLS (b is for big), a MATLAB software library that allows to use RLS on very large
 matrices by means of memory-mapped storage and a simple distributed task manager.

-GURLS++, a C++ standalone implementation of GURLS, with additional simple API's for specific learning pipelines

-bGURLS++, a  C++ standalone implementation of bGURLS.

Documentation

=============

- Webpage 
	the GURLS webpage can be found at: http://lcsl.mit.edu/#/downloads/gurls

- Reference paper
	Tacchetti,  Mallapragada, Santoro and Rosasco,
	Gurls: a least squares-based library for state of the art supervised learning
	accepted for publication on JMLR

- Installation instructions can be found here:
	https://github.com/LCSL/GURLS/wiki/2-Getting-Started

- Quick intructions on how to run the libraries for a default case can be found here: 
	https://github.com/LCSL/GURLS/wiki/2-Getting-Started#wiki-Hello_World

- A User manual with several examples can be found here:
	https://github.com/LCSL/GURLS/wiki/3-User-Manual#wiki-User_Manual

- A collection of the most useful and common pipelines can be found here:
	https://github.com/LCSL/GURLS/wiki/3-User-Manual#wiki-Examples

- The list of all the available methods of the libraries can be found at
	https://github.com/LCSL/GURLS/wiki/4-Available-methods

- C++ Code Documentation can be found at: http://lcsl.github.io/GURLS/

- Further Documentation
	* Have a look at the README files of each individual package.
	
	* In gurls-manual.pdf you can find both the installation instructions 
	  and user manual, together with the Matlab and C++ Developer's Guide. 
	  GURLS is designed for easy expansion. Give it a try!
	
	* In recursiveRLS-tutorial.pdf you can find a simple Tutorial for the Recursive 
	  RLS API 

	* The description of the available methods, demos and data 
	  for each package can be found at
	 	https://github.com/CBCL/GURLS/wiki/4-Code-Description