weka.core.stats

Class NumericAttributeBinData

java.lang.Object

weka.core.stats.NumericAttributeBinData

All Implemented Interfaces:

weka.core.Aggregateable<NumericAttributeBinData>

public class NumericAttributeBinData
extends java.lang.Object
implements weka.core.Aggregateable<NumericAttributeBinData>

Class for managing bin data for a histogram based on an attribute

Version:

$Revision: 13233 $

Author:

Mark Hall (mhall{[at]}pentaho{[dot]}com)

Field Summary

Fields

Modifier and Type	Field and Description
static int	MAX_BINS Maximum bins to create

Constructor Summary

Constructors

Constructor and Description
NumericAttributeBinData(java.lang.String attName, weka.core.Attribute summaryAtt, int maxBins) Constructor
NumericAttributeBinData(java.lang.String attName, double numPoints, double min, double max, double stdDev, double missingFreq, double lowerPercentile, double upperPercentile, int maxBins) Constructor
NumericAttributeBinData(java.lang.String attName, double numPoints, double min, double max, double stdDev, double missingFreq, int maxBins) Constructor

Method Summary

Modifier and Type	Method and Description
void	addValue(double value, double weight) Add a value to the histogram.
NumericAttributeBinData	aggregate(NumericAttributeBinData b)
void	finalizeAggregation()
java.lang.String	getAttributeName() Get the name of the attribute that this histogram is for
java.util.List<java.lang.Double>	getBinFreqs() Get a list of bin frequencies for this histogram
java.util.List<java.lang.String>	getBinLabels() Get a list of bin labels for this histogram
double	getBinWidth() Get the bin width for this attribute
double	getMissingFreq() Get the number of missing values
int	getNumBins() Get the number of bins for this attribute
static void	main(java.lang.String[] args)
static int	numBinsHeuristic(double stdDev, double numPoints, double min, double max, int maxBins) Compute the number of bins for a histogram given summary stats
java.lang.String	toString()

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

MAX_BINS

public static final int MAX_BINS

Maximum bins to create

See Also:

Constant Field Values

Constructor Detail

NumericAttributeBinData

public NumericAttributeBinData(java.lang.String attName,
                               weka.core.Attribute summaryAtt,
                               int maxBins)

Constructor

Parameters:

attName - the name of the attribute

summaryAtt - the summary attribute containing the bin cutpoints and frequencies

maxBins - the maximum number of bins to allow (if setting bin cutpoints based on range and overall count) or -1 to use the default max.

NumericAttributeBinData

public NumericAttributeBinData(java.lang.String attName,
                               double numPoints,
                               double min,
                               double max,
                               double stdDev,
                               double missingFreq,
                               int maxBins)

Constructor

Parameters:

attName - the name of the attribute

numPoints - the number of points that have been seen for this attribute

min - the minimum value

max - the maximum value

stdDev - the standard deviation

missingFreq - the number of missing values

maxBins - the maximum number of bins to allow (if setting bin cutpoints based on range and overall count) or -1 to use the default max.

NumericAttributeBinData

public NumericAttributeBinData(java.lang.String attName,
                               double numPoints,
                               double min,
                               double max,
                               double stdDev,
                               double missingFreq,
                               double lowerPercentile,
                               double upperPercentile,
                               int maxBins)

Constructor

Parameters:

attName - the name of the attribute

numPoints - the number of points that have been seen for this attribute

min - the minimum value

max - the maximum value

stdDev - the standard deviation

missingFreq - the number of missing values

lowerPercentile - the lower percentile to use for the equal width binning

upperPercentile - the upper percentile to use for hte equal width binning

maxBins - the maximum number of bins to use

Method Detail

getNumBins

public int getNumBins()

Get the number of bins for this attribute

Returns:

the number of bins

getBinWidth

public double getBinWidth()

Get the bin width for this attribute

Returns:

the bin width for this attribute

getBinLabels

public java.util.List<java.lang.String> getBinLabels()

Get a list of bin labels for this histogram

Returns:

a list of bin labels

getBinFreqs

public java.util.List<java.lang.Double> getBinFreqs()

Get a list of bin frequencies for this histogram

Returns:

a list of bin frequencies

getMissingFreq

public double getMissingFreq()

Get the number of missing values

Returns:

the number of missing values

getAttributeName

public java.lang.String getAttributeName()

Get the name of the attribute that this histogram is for

Returns:

the name of the attribute that this histogram is for

addValue

public void addValue(double value,
                     double weight)

Add a value to the histogram. Finds the correct bin and increases the frequency

Parameters:

value - the value to add

weight - the weight

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

aggregate

public NumericAttributeBinData aggregate(NumericAttributeBinData b)
                                  throws java.lang.Exception

Specified by:

aggregate in interface weka.core.Aggregateable<NumericAttributeBinData>

Throws:

java.lang.Exception

finalizeAggregation

public void finalizeAggregation()
                         throws java.lang.Exception

Specified by:

finalizeAggregation in interface weka.core.Aggregateable<NumericAttributeBinData>

Throws:

java.lang.Exception

numBinsHeuristic

public static int numBinsHeuristic(double stdDev,
                                   double numPoints,
                                   double min,
                                   double max,
                                   int maxBins)

Compute the number of bins for a histogram given summary stats

Parameters:

stdDev - the standard deviation of the variable in question

numPoints - the number of observed data points

min - the minimum value

max - the maximum

maxBins - the maximum number of bins to allow

Returns:

the number of bins

main

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