weka.classifiers
Class BVDecompose

java.lang.Object
  weka.classifiers.BVDecompose

All Implemented Interfaces:

OptionHandler

public class BVDecompose

extends java.lang.Object

implements OptionHandler

Class for performing a Bias-Variance decomposition on any classifier using the method specified in:

R. Kohavi & D. Wolpert (1996), Bias plus variance decomposition for zero-one loss functions, in Proc. of the Thirteenth International Machine Learning Conference (ICML96) download postscript.

Valid options are:

-D
Turn on debugging output.

-W classname
Specify the full class name of a learner to perform the decomposition on (required).

-t filename
Set the arff file to use for the decomposition (required).

-T num
Specify the number of instances in the training pool (default 100).

-c num
Specify the index of the class attribute (default last).

-x num
Set the number of train iterations (default 50).

-s num
Set the seed for the dataset randomisation (default 1).

Options after -- are passed to the designated sub-learner.

Version:

$Revision: 1.8 $

Author:

Len Trigg (trigg@cs.waikato.ac.nz)

Field Summary

protected double	m_Bias The calculated bias (squared)
protected Classifier	m_Classifier An instantiated base classifier used for getting and testing options.
protected java.lang.String[]	m_ClassifierOptions The options to be passed to the base classifier.
protected int	m_ClassIndex The index of the class attribute
protected java.lang.String	m_DataFileName The name of the data file used for the decomposition
protected boolean	m_Debug Debugging mode, gives extra output if true
protected double	m_Error The error rate
protected int	m_Seed The random number seed
protected double	m_Sigma The calculated sigma (squared)
protected int	m_TrainIterations The number of train iterations
protected int	m_TrainPoolSize The number of instances used in the training pool
protected double	m_Variance The calculated variance

Constructor Summary

BVDecompose()

Method Summary

void	decompose() Carry out the bias-variance decomposition
double	getBias() Get the calculated bias squared
Classifier	getClassifier() Gets the name of the classifier being analysed
int	getClassIndex() Get the index (starting from 1) of the attribute used as the class.
java.lang.String	getDataFileName() Get the name of the data file used for the decomposition
boolean	getDebug() Gets whether debugging is turned on
double	getError() Get the calculated error rate
java.lang.String[]	getOptions() Gets the current settings of the CheckClassifier.
int	getSeed() Gets the random number seed
double	getSigma() Get the calculated sigma squared
int	getTrainIterations() Gets the maximum number of boost iterations
int	getTrainPoolSize() Get the number of instances in the training pool.
double	getVariance() Get the calculated variance
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] args) Test method for this class
void	setClassifier(Classifier newClassifier) Set the classifiers being analysed
void	setClassIndex(int classIndex) Sets index of attribute to discretize on
void	setDataFileName(java.lang.String dataFileName) Sets the maximum number of boost iterations
void	setDebug(boolean debug) Sets debugging mode
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setSeed(int seed) Sets the random number seed
void	setTrainIterations(int trainIterations) Sets the maximum number of boost iterations
void	setTrainPoolSize(int numTrain) Set the number of instances in the training pool.
java.lang.String	toString() Returns description of the bias-variance decomposition results.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

m_Debug

protected boolean m_Debug

Debugging mode, gives extra output if true

m_Classifier

protected Classifier m_Classifier

An instantiated base classifier used for getting and testing options.

m_ClassifierOptions

protected java.lang.String[] m_ClassifierOptions

The options to be passed to the base classifier.

m_TrainIterations

protected int m_TrainIterations

The number of train iterations

m_DataFileName

protected java.lang.String m_DataFileName

The name of the data file used for the decomposition

m_ClassIndex

protected int m_ClassIndex

The index of the class attribute

m_Seed

protected int m_Seed

The random number seed

m_Bias

protected double m_Bias

The calculated bias (squared)

m_Variance

protected double m_Variance

The calculated variance

m_Sigma

protected double m_Sigma

The calculated sigma (squared)

m_Error

protected double m_Error

The error rate

m_TrainPoolSize

protected int m_TrainPoolSize

The number of instances used in the training pool

Constructor Detail

BVDecompose

public BVDecompose()

Method Detail

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Returns:

an enumeration of all the available options.

setOptions

public void setOptions(java.lang.String[] options)
                throws java.lang.Exception

Parses a given list of options. Valid options are:

-D
Turn on debugging output.

-W classname
Specify the full class name of a learner to perform the decomposition on (required).

-t filename
Set the arff file to use for the decomposition (required).

-T num
Specify the number of instances in the training pool (default 100).

-c num
Specify the index of the class attribute (default last).

-x num
Set the number of train iterations (default 50).

-s num
Set the seed for the dataset randomisation (default 1).

Options after -- are passed to the designated sub-learner.

Specified by:

setOptions in interface OptionHandler

Parameters:

options - the list of options as an array of strings

Throws:

java.lang.Exception - if an option is not supported

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the CheckClassifier.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions

getTrainPoolSize

public int getTrainPoolSize()

Get the number of instances in the training pool.

Returns:

number of instances in the training pool.

setTrainPoolSize

public void setTrainPoolSize(int numTrain)

Set the number of instances in the training pool.

Parameters:

numTrain - number of instances in the training pool.

setClassifier

public void setClassifier(Classifier newClassifier)

Set the classifiers being analysed

Parameters:

newClassifier - the Classifier to use.

getClassifier

public Classifier getClassifier()

Gets the name of the classifier being analysed

Returns:

the classifier being analysed.

setDebug

public void setDebug(boolean debug)

Sets debugging mode

Parameters:

debug - true if debug output should be printed

getDebug

public boolean getDebug()

Gets whether debugging is turned on

Returns:

true if debugging output is on

setSeed

public void setSeed(int seed)

Sets the random number seed

getSeed

public int getSeed()

Gets the random number seed

Returns:

the random number seed

setTrainIterations

public void setTrainIterations(int trainIterations)

Sets the maximum number of boost iterations

getTrainIterations

public int getTrainIterations()

Gets the maximum number of boost iterations

Returns:

the maximum number of boost iterations

setDataFileName

public void setDataFileName(java.lang.String dataFileName)

Sets the maximum number of boost iterations

getDataFileName

public java.lang.String getDataFileName()

Get the name of the data file used for the decomposition

Returns:

the name of the data file

getClassIndex

public int getClassIndex()

Get the index (starting from 1) of the attribute used as the class.

Returns:

the index of the class attribute

setClassIndex

public void setClassIndex(int classIndex)

Sets index of attribute to discretize on

getBias

public double getBias()

Get the calculated bias squared

Returns:

the bias squared

getVariance

public double getVariance()

Get the calculated variance

Returns:

the variance

getSigma

public double getSigma()

Get the calculated sigma squared

Returns:

the sigma squared

getError

public double getError()

Get the calculated error rate

Returns:

the error rate

decompose

public void decompose()
               throws java.lang.Exception

Carry out the bias-variance decomposition

Throws:

java.lang.Exception - if the decomposition couldn't be carried out

toString

public java.lang.String toString()

Returns description of the bias-variance decomposition results.

Returns:

the bias-variance decomposition results as a string

main

public static void main(java.lang.String[] args)

Test method for this class

Parameters:

args - the command line arguments