readme.txt

AdaBoost is an efficient tool in machine learning. It can combine a series of weak learners into a s

-------------------------------------------------
--- Document for MATLAB interface of Boosting ---
-------------------------------------------------

Table of Contents
=================
- Introduction
- Installation
- Usage
- Examples

Introduction
============
This source code is freely available for non-commercial use such as academic research and education. For other purpose, please contact me: [e-mail]

AdaBoost is an efficient tool in machine learning. It can combine a series of weak learners into a strong learner. Besides pattern classification, it also can be applied into feature selection. This document explains the use of AdaBoost.

This tool provides a simple interface to Boosting, a library for Multi-Class AdaBoost (http://njustyw.googlepages.com/). 
It is very easy to use as the usage and the way of specifying parameters are the same as that of Boosting.

Installation
============

On Windows systems, pre-built 'boosttrain.dll' and 'boostpredict.dll' are
included in this package, so no need to conduct installation. If you
have modified the sources and would like to re-build the package, type
'mex -setup' in MATLAB to choose a compiler for mex first. Then type
'make' to start the installation.

Example:
        matlab> mex -setup
        (ps: MATLAB will show the following messages to setup default compiler.)
        Please choose your compiler for building external interface (MEX) files: 
        Would you like mex to locate installed compilers [y]/n? y
        Select a compiler: 
        [1] Microsoft Visual C/C++ version 6.0 in C:\Program Files\Microsoft Visual Studio 
        [0] None 
        Compiler: 1
        Please verify your choices: 
        Compiler: Microsoft Visual C/C++ 6.0 
        Location: C:\Program Files\Microsoft Visual Studio 
        Are these correct?([y]/n): y

        matlab> make

For list of supported/compatible compilers for MATLAB, please check the
following page:

http://www.mathworks.com/support/tech-notes/1600/1601.html

Usage
=====

matlab> model = boosttrain(train_data, train_label, iteration);

        -training_data:
            An m by n matrix of m training instances with n features.
        -training_label:
            An m by 1 vector of training labels.
        -iterations:
            step number of iteration for AdaBoost(default 100).


matlab> predicted_label = boostpredict(model, predict_data);

        -model:
            The output of boosttrain.
        -predict_data:
            An m by n matrix of m predicting instances with n features.


matlab> [predicted_label, accuracy] = boostpredict(model, test_data, test_label);

        -model:
            The output of boosttrain.
        -test_data:
            An m by n matrix of m testing instances with n features.
        -test_label:
            An m by 1 vector of testing labels.
        



Examples
========

Train and test on the data:

matlab> load data.mat
matlab> model = boosttrain(train_data, train_label, 100); %training
matlab> [predict_label, accuracy] = boostpredict(model, train_data, train_label); % test the training data
matlab> [predict_label, accuracy] = boostpredict(model, test_data, test_label); % test the testing data
matlab> predict_label = boostpredict(model, test_data); % predict the testing data




=================
Copyright (c) 2006-2008, Great Yao.  Email:[e-mail]

http://njustyw.googlepages.com/