weka.classifiers.functions
Class SimpleLogistic

java.lang.Object
  weka.classifiers.Classifier
      weka.classifiers.functions.SimpleLogistic

All Implemented Interfaces:

AdditionalMeasureProducer, java.lang.Cloneable, OptionHandler, java.io.Serializable, WeightedInstancesHandler

public class SimpleLogistic

extends Classifier

implements OptionHandler, AdditionalMeasureProducer, WeightedInstancesHandler

Class for building a logistic regression model using LogitBoost. Incorporates attribute selection by fitting simple regression functions in LogitBoost. For more information, see master thesis "Logistic Model Trees" (Niels Landwehr, 2003)

Valid options are:

-I iterations
Set fixed number of iterations for LogitBoost (instead of using cross-validation).

-S
Select the number of LogitBoost iterations that gives minimal error on the training set (instead of using cross-validation).

-P
Minimize error on probabilities instead of misclassification error.

-M iterations
Set maximum number of iterations for LogitBoost.

-H iter
Set parameter for heuristic for early stopping of LogitBoost. If enabled, the minimum is selected greedily, stopping if the current minimum has not changed for iter iterations. By default, heuristic is enabled with value 50. Set to zero to disable heuristic.

Version:

$Revision: 1.4 $

Author:

Niels Landwehr

See Also:

Serialized Form

Field Summary

protected LogisticBase	m_boostedModel The actual logistic regression model
protected boolean	m_errorOnProbabilities If true, use minimize error on probabilities instead of misclassification error
protected int	m_heuristicStop Parameter for the heuristic for early stopping of LogitBoost
protected int	m_maxBoostingIterations Maximum number of iterations for LogitBoost
protected NominalToBinary	m_NominalToBinary Filter for converting nominal attributes to binary ones
protected int	m_numBoostingIterations If non-negative, use this as fixed number of LogitBoost iterations
protected ReplaceMissingValues	m_ReplaceMissingValues Filter for replacing missing values
protected boolean	m_useCrossValidation If true, cross-validate number of LogitBoost iterations

Fields inherited from class weka.classifiers.Classifier

m_Debug

Constructor Summary

SimpleLogistic()
Constructor for creating SimpleLogistic object with standard options.

SimpleLogistic(int numBoostingIterations, boolean useCrossValidation, boolean errorOnProbabilities)
Constructor for creating SimpleLogistic object.

Method Summary

void	buildClassifier(Instances data) Builds the logistic regression using LogitBoost.
double[]	distributionForInstance(Instance inst) Returns class probabilities for an instance.
java.util.Enumeration	enumerateMeasures() Returns an enumeration of the additional measure names
java.lang.String	errorOnProbabilitiesTipText() Returns the tip text for this property
boolean	getErrorOnProbabilities() Get the value of errorOnProbabilities.
int	getHeuristicStop() Get the value of heuristicStop.
int	getMaxBoostingIterations() Get the value of maxBoostingIterations.
double	getMeasure(java.lang.String additionalMeasureName) Returns the value of the named measure
int	getNumBoostingIterations() Get the value of numBoostingIterations.
int	getNumRegressions() Get the number of LogitBoost iterations performed (= the number of regression functions fit by LogitBoost).
java.lang.String[]	getOptions() Gets the current settings of the Classifier.
boolean	getUseCrossValidation() Get the value of useCrossValidation.
java.lang.String	globalInfo() Returns a string describing classifier
java.lang.String	heuristicStopTipText() Returns the tip text for this property
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class
java.lang.String	maxBoostingIterationsTipText() Returns the tip text for this property
double	measureAttributesUsed() Returns the fraction of all attributes in the data that are used in the logistic model (in percent).
java.lang.String	numBoostingIterationsTipText() Returns the tip text for this property
void	setErrorOnProbabilities(boolean l) Set the value of errorOnProbabilities.
void	setHeuristicStop(int n) Set the value of heuristicStop.
void	setMaxBoostingIterations(int n) Set the value of maxBoostingIterations.
void	setNumBoostingIterations(int n) Set the value of numBoostingIterations.
void	setOptions(java.lang.String[] options) Parses a given list of options.
void	setUseCrossValidation(boolean l) Set the value of useCrossValidation.
java.lang.String	toString() Returns a description of the logistic model (attributes/coefficients).
java.lang.String	useCrossValidationTipText() Returns the tip text for this property

Methods inherited from class weka.classifiers.Classifier

classifyInstance, debugTipText, forName, getDebug, makeCopies, setDebug

Methods inherited from class java.lang.Object

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

Field Detail

m_boostedModel

protected LogisticBase m_boostedModel

The actual logistic regression model

m_NominalToBinary

protected NominalToBinary m_NominalToBinary

Filter for converting nominal attributes to binary ones

m_ReplaceMissingValues

protected ReplaceMissingValues m_ReplaceMissingValues

Filter for replacing missing values

m_numBoostingIterations

protected int m_numBoostingIterations

If non-negative, use this as fixed number of LogitBoost iterations

m_maxBoostingIterations

protected int m_maxBoostingIterations

Maximum number of iterations for LogitBoost

m_heuristicStop

protected int m_heuristicStop

Parameter for the heuristic for early stopping of LogitBoost

m_useCrossValidation

protected boolean m_useCrossValidation

If true, cross-validate number of LogitBoost iterations

m_errorOnProbabilities

protected boolean m_errorOnProbabilities

If true, use minimize error on probabilities instead of misclassification error

Constructor Detail

SimpleLogistic

public SimpleLogistic()

Constructor for creating SimpleLogistic object with standard options.

SimpleLogistic

public SimpleLogistic(int numBoostingIterations,
                      boolean useCrossValidation,
                      boolean errorOnProbabilities)

Constructor for creating SimpleLogistic object.

Parameters:

numBoostingIterations - if non-negative, use this as fixed number of iterations for LogitBoost

useCrossValidation - cross-validate number of LogitBoost iterations.

errorOnProbabilities - minimize error on probabilities instead of misclassification error

Method Detail

buildClassifier

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

Builds the logistic regression using LogitBoost.

Specified by:

buildClassifier in class Classifier

Parameters:

data - the training data

Throws:

java.lang.Exception - if something goes wrong

distributionForInstance

public double[] distributionForInstance(Instance inst)
                                 throws java.lang.Exception

Returns class probabilities for an instance.

Overrides:

distributionForInstance in class Classifier

Parameters:

inst - the instance to be classified

Returns:

an array containing the estimated membership probabilities of the test instance in each class or the numeric prediction

Throws:

java.lang.Exception - if distribution can't be computed successfully

listOptions

public java.util.Enumeration listOptions()

Returns an enumeration describing the available options.

Specified by:

listOptions in interface OptionHandler

Overrides:

listOptions in class Classifier

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.

Specified by:

setOptions in interface OptionHandler

Overrides:

setOptions in class Classifier

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 Classifier.

Specified by:

getOptions in interface OptionHandler

Overrides:

getOptions in class Classifier

Returns:

an array of strings suitable for passing to setOptions

getNumBoostingIterations

public int getNumBoostingIterations()

Get the value of numBoostingIterations.

getUseCrossValidation

public boolean getUseCrossValidation()

Get the value of useCrossValidation.

getErrorOnProbabilities

public boolean getErrorOnProbabilities()

Get the value of errorOnProbabilities.

getMaxBoostingIterations

public int getMaxBoostingIterations()

Get the value of maxBoostingIterations.

getHeuristicStop

public int getHeuristicStop()

Get the value of heuristicStop.

setNumBoostingIterations

public void setNumBoostingIterations(int n)

Set the value of numBoostingIterations.

setUseCrossValidation

public void setUseCrossValidation(boolean l)

Set the value of useCrossValidation.

setErrorOnProbabilities

public void setErrorOnProbabilities(boolean l)

Set the value of errorOnProbabilities.

setMaxBoostingIterations

public void setMaxBoostingIterations(int n)

Set the value of maxBoostingIterations.

setHeuristicStop

public void setHeuristicStop(int n)

Set the value of heuristicStop.

getNumRegressions

public int getNumRegressions()

Get the number of LogitBoost iterations performed (= the number of regression functions fit by LogitBoost).

toString

public java.lang.String toString()

Returns a description of the logistic model (attributes/coefficients).

measureAttributesUsed

public double measureAttributesUsed()

Returns the fraction of all attributes in the data that are used in the logistic model (in percent). An attribute is used in the model if it is used in any of the models for the different classes.

enumerateMeasures

public java.util.Enumeration enumerateMeasures()

Returns an enumeration of the additional measure names

Specified by:

enumerateMeasures in interface AdditionalMeasureProducer

Returns:

an enumeration of the measure names

getMeasure

public double getMeasure(java.lang.String additionalMeasureName)

Returns the value of the named measure

Specified by:

getMeasure in interface AdditionalMeasureProducer

Parameters:

additionalMeasureName - the name of the measure to query for its value

Returns:

the value of the named measure

Throws:

java.lang.IllegalArgumentException - if the named measure is not supported

globalInfo

public java.lang.String globalInfo()

Returns a string describing classifier

Returns:

a description suitable for displaying in the explorer/experimenter gui

numBoostingIterationsTipText

public java.lang.String numBoostingIterationsTipText()

Returns the tip text for this property

Returns:

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

useCrossValidationTipText

public java.lang.String useCrossValidationTipText()

Returns the tip text for this property

Returns:

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

errorOnProbabilitiesTipText

public java.lang.String errorOnProbabilitiesTipText()

Returns the tip text for this property

Returns:

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

maxBoostingIterationsTipText

public java.lang.String maxBoostingIterationsTipText()

Returns the tip text for this property

Returns:

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

heuristicStopTipText

public java.lang.String heuristicStopTipText()

Returns the tip text for this property

Returns:

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

main

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

Main method for testing this class