weka.attributeSelection
Class WrapperSubsetEval

java.lang.Object
  weka.attributeSelection.ASEvaluation
      weka.attributeSelection.SubsetEvaluator
          weka.attributeSelection.WrapperSubsetEval

All Implemented Interfaces:

OptionHandler, java.io.Serializable

public class WrapperSubsetEval

extends SubsetEvaluator

implements OptionHandler

Wrapper attribute subset evaluator.

For more information see:
Kohavi, R., John G., Wrappers for Feature Subset Selection. In Artificial Intelligence journal, special issue on relevance, Vol. 97, Nos 1-2, pp.273-324.

Valid options are:

-B
Class name of base learner to use for accuracy estimation. Place any classifier options last on the command line following a "--". Eg -B weka.classifiers.bayes.NaiveBayes ... -- -K

-F
Number of cross validation folds to use for estimating accuracy.

-T
Threshold by which to execute another cross validation (standard deviation ---expressed as a percentage of the mean).

-R
Seed for cross validation accuracy estimation. (default = 1)

Version:

$Revision: 1.21 $

Author:

Mark Hall (mhall@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

private Classifier	m_BaseClassifier holds the base classifier object
private int	m_classIndex class index
private Evaluation	m_Evaluation holds an evaluation object
private int	m_folds number of folds to use for cross validation
private int	m_numAttribs number of attributes in the training data
private int	m_numInstances number of instances in the training data
private int	m_seed random number seed
private double	m_threshold the threshold by which to do further cross validations when estimating the accuracy of a subset
private Instances	m_trainInstances training instances

Constructor Summary

WrapperSubsetEval()
Constructor.

Method Summary

void	buildEvaluator(Instances data) Generates a attribute evaluator.
java.lang.String	classifierTipText() Returns the tip text for this property
double	evaluateSubset(java.util.BitSet subset) Evaluates a subset of attributes
java.lang.String	foldsTipText() Returns the tip text for this property
Classifier	getClassifier() Get the classifier used as the base learner.
int	getFolds() Get the number of folds used for accuracy estimation
java.lang.String[]	getOptions() Gets the current settings of WrapperSubsetEval.
int	getSeed() Get the random number seed used for cross validation
double	getThreshold() Get the value of the threshold
java.lang.String	globalInfo() Returns a string describing this attribute evaluator
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] args) Main method for testing this class.
private boolean	repeat(double[] repError, int entries) decides whether to do another repeat of cross validation.
protected void	resetOptions()
java.lang.String	seedTipText() Returns the tip text for this property
void	setClassifier(Classifier newClassifier) Set the classifier to use for accuracy estimation
void	setFolds(int f) Set the number of folds to use for accuracy estimation
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setSeed(int s) Set the seed to use for cross validation
void	setThreshold(double t) Set the value of the threshold for repeating cross validation
java.lang.String	thresholdTipText() Returns the tip text for this property
java.lang.String	toString() Returns a string describing the wrapper

Methods inherited from class weka.attributeSelection.ASEvaluation

forName, makeCopies, postProcess

Methods inherited from class java.lang.Object

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

Field Detail

m_trainInstances

private Instances m_trainInstances

training instances

m_classIndex

private int m_classIndex

class index

m_numAttribs

private int m_numAttribs

number of attributes in the training data

m_numInstances

private int m_numInstances

number of instances in the training data

m_Evaluation

private Evaluation m_Evaluation

holds an evaluation object

m_BaseClassifier

private Classifier m_BaseClassifier

holds the base classifier object

m_folds

private int m_folds

number of folds to use for cross validation

m_seed

private int m_seed

random number seed

m_threshold

private double m_threshold

the threshold by which to do further cross validations when estimating the accuracy of a subset

Constructor Detail

WrapperSubsetEval

public WrapperSubsetEval()

Constructor. Calls restOptions to set default options

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing this attribute evaluator

Returns:

a description of the evaluator suitable for displaying in the explorer/experimenter gui

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:

-B
Class name of base learner to use for accuracy estimation. Place any classifier options last on the command line following a "--". Eg -B weka.classifiers.bayes.NaiveBayes ... -- -K

-F
Number of cross validation folds to use for estimating accuracy.

-T
Threshold by which to execute another cross validation (standard deviation ---expressed as a percentage of the mean).

-R
Seed for cross validation accuracy estimation. (default = 1)

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

thresholdTipText

public java.lang.String thresholdTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setThreshold

public void setThreshold(double t)

Set the value of the threshold for repeating cross validation

Parameters:

t - the value of the threshold

getThreshold

public double getThreshold()

Get the value of the threshold

Returns:

the threshold as a double

foldsTipText

public java.lang.String foldsTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setFolds

public void setFolds(int f)

Set the number of folds to use for accuracy estimation

Parameters:

f - the number of folds

getFolds

public int getFolds()

Get the number of folds used for accuracy estimation

Returns:

the number of folds

seedTipText

public java.lang.String seedTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setSeed

public void setSeed(int s)

Set the seed to use for cross validation

Parameters:

s - the seed

getSeed

public int getSeed()

Get the random number seed used for cross validation

Returns:

the seed

classifierTipText

public java.lang.String classifierTipText()

Returns the tip text for this property

Returns:

tip text for this property suitable for displaying in the explorer/experimenter gui

setClassifier

public void setClassifier(Classifier newClassifier)

Set the classifier to use for accuracy estimation

Parameters:

newClassifier - the Classifier to use.

getClassifier

public Classifier getClassifier()

Get the classifier used as the base learner.

Returns:

the classifier used as the classifier

getOptions

public java.lang.String[] getOptions()

Gets the current settings of WrapperSubsetEval.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions()

resetOptions

protected void resetOptions()

buildEvaluator

public void buildEvaluator(Instances data)
                    throws java.lang.Exception

Generates a attribute evaluator. Has to initialize all fields of the evaluator that are not being set via options.

Specified by:

buildEvaluator in class ASEvaluation

Parameters:

data - set of instances serving as training data

Throws:

java.lang.Exception - if the evaluator has not been generated successfully

evaluateSubset

public double evaluateSubset(java.util.BitSet subset)
                      throws java.lang.Exception

Evaluates a subset of attributes

Specified by:

evaluateSubset in class SubsetEvaluator

Parameters:

subset - a bitset representing the attribute subset to be evaluated

Returns:

the "merit" of the subset

Throws:

java.lang.Exception - if the subset could not be evaluated

toString

public java.lang.String toString()

Returns a string describing the wrapper

Returns:

the description as a string

repeat

private boolean repeat(double[] repError,
                       int entries)

decides whether to do another repeat of cross validation. If the standard deviation of the cross validations is greater than threshold% of the mean (default 1%) then another repeat is done.

Parameters:

repError - an array of cross validation results

entries - the number of cross validations done so far

Returns:

true if another cv is to be done

main

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

Main method for testing this class.

Parameters:

args - the options