weka.classifiers.trees.ht

Class GaussianConditionalSufficientStats

java.lang.Object

weka.classifiers.trees.ht.ConditionalSufficientStats

weka.classifiers.trees.ht.GaussianConditionalSufficientStats

All Implemented Interfaces:

java.io.Serializable

public class GaussianConditionalSufficientStats
extends ConditionalSufficientStats
implements java.io.Serializable

Maintains sufficient stats for a Gaussian distribution for a numeric attribute

See Also:

Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
protected class	GaussianConditionalSufficientStats.GaussianEstimator Inner class that implements a Gaussian estimator

Field Summary

Fields

Modifier and Type	Field and Description
protected java.util.Map<java.lang.String,java.lang.Double>	m_maxValObservedPerClass
protected java.util.Map<java.lang.String,java.lang.Double>	m_minValObservedPerClass
protected int	m_numBins

Fields inherited from class weka.classifiers.trees.ht.ConditionalSufficientStats

m_classLookup

Constructor Summary

Constructors

Constructor and Description
GaussianConditionalSufficientStats()

Method Summary

Modifier and Type	Method and Description
SplitCandidate	bestSplit(SplitMetric splitMetric, java.util.Map<java.lang.String,WeightMass> preSplitDist, java.lang.String attName) Return the best split
protected java.util.List<java.util.Map<java.lang.String,WeightMass>>	classDistsAfterSplit(double splitVal)
int	getNumBins()
protected java.util.TreeSet<java.lang.Double>	getSplitPointCandidates()
double	probabilityOfAttValConditionedOnClass(double attVal, java.lang.String classVal) Return the probability of an attribute value conditioned on a class value
void	setNumBins(int b)
void	update(double attVal, java.lang.String classVal, double weight) Update this stat with the supplied attribute value and class value

Methods inherited from class java.lang.Object

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

Field Detail

m_minValObservedPerClass

protected java.util.Map<java.lang.String,java.lang.Double> m_minValObservedPerClass

m_maxValObservedPerClass

protected java.util.Map<java.lang.String,java.lang.Double> m_maxValObservedPerClass

m_numBins

protected int m_numBins

Constructor Detail

GaussianConditionalSufficientStats

public GaussianConditionalSufficientStats()

Method Detail

setNumBins

public void setNumBins(int b)

getNumBins

public int getNumBins()

update

public void update(double attVal,
                   java.lang.String classVal,
                   double weight)

Description copied from class: ConditionalSufficientStats

Update this stat with the supplied attribute value and class value

Specified by:

update in class ConditionalSufficientStats

Parameters:

attVal - the value of the attribute

classVal - the class value

weight - the weight of this observation

probabilityOfAttValConditionedOnClass

public double probabilityOfAttValConditionedOnClass(double attVal,
                                                    java.lang.String classVal)

Description copied from class: ConditionalSufficientStats

Return the probability of an attribute value conditioned on a class value

Specified by:

probabilityOfAttValConditionedOnClass in class ConditionalSufficientStats

Parameters:

attVal - the attribute value to compute the conditional probability for

classVal - the class value

Returns:

the probability

getSplitPointCandidates

protected java.util.TreeSet<java.lang.Double> getSplitPointCandidates()

classDistsAfterSplit

protected java.util.List<java.util.Map<java.lang.String,WeightMass>> classDistsAfterSplit(double splitVal)

bestSplit

public SplitCandidate bestSplit(SplitMetric splitMetric,
                                java.util.Map<java.lang.String,WeightMass> preSplitDist,
                                java.lang.String attName)

Description copied from class: ConditionalSufficientStats

Return the best split

Specified by:

bestSplit in class ConditionalSufficientStats

Parameters:

splitMetric - the split metric to use

preSplitDist - the distribution of class values prior to splitting

attName - the name of the attribute being considered for splitting

Returns:

the best split for the attribute