weka.estimators

Class EstimatorUtils

java.lang.Object

weka.estimators.EstimatorUtils

All Implemented Interfaces:

RevisionHandler

public class EstimatorUtils
extends java.lang.Object
implements RevisionHandler

Contains static utility functions for Estimators.

Version:

$Revision: 15521 $

Author:

Gabi Schmidberger (gabi@cs.waikato.ac.nz)

Constructor Summary

Constructors

Constructor and Description
EstimatorUtils()

Method Summary

Modifier and Type	Method and Description
static java.lang.String	cutpointsToString(double[] cutPoints, boolean[] cutAndLeft) Returns a string representing the cutpoints.
static double	findMinDistance(Instances inst, int attrIndex) Find the minimum distance between values.
static Instances	getInstancesFromClass(Instances data, int classIndex, double classValue) Returns a dataset that contains of all instances of a certain class value.
static java.util.Vector<java.lang.Object>	getInstancesFromClass(Instances data, int attrIndex, int classIndex, double classValue, Instances workData) Returns a dataset that contains all instances of a certain class value.
static Instances	getInstancesFromValue(Instances data, int index, double v) Returns a dataset that contains of all instances of a certain value for the given attribute.
static int	getMinMax(Instances inst, int attrIndex, double[] minMax) Find the minimum and the maximum of the attribute and return it in the last parameter.
java.lang.String	getRevision() Returns the revision string.
static void	writeCurve(java.lang.String f, Estimator est, double min, double max, int numPoints) Output of an n points of a density curve.
static void	writeCurve(java.lang.String f, Estimator est, Estimator classEst, double classIndex, double min, double max, int numPoints) Output of an n points of a density curve.

Methods inherited from class java.lang.Object

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

Constructor Detail

EstimatorUtils

public EstimatorUtils()

Method Detail

findMinDistance

public static double findMinDistance(Instances inst,
                                     int attrIndex)

Find the minimum distance between values. Data is assumed to be sorted based on the given attribute. Missing values are skipped if they are at the beginning or the end of the data.

Parameters:

inst - sorted instances, sorted

attrIndex - index of the attribute, they are sorted after

Returns:

the minimal distance

getMinMax

public static int getMinMax(Instances inst,
                            int attrIndex,
                            double[] minMax)
                     throws java.lang.Exception

Find the minimum and the maximum of the attribute and return it in the last parameter.

Parameters:

inst - instances used to build the estimator

attrIndex - index of the attribute

minMax - the array to return minimum and maximum in

Returns:

number of not missing values

Throws:

java.lang.Exception - if parameter minMax wasn't initialized properly

getInstancesFromClass

public static java.util.Vector<java.lang.Object> getInstancesFromClass(Instances data,
                                                                       int attrIndex,
                                                                       int classIndex,
                                                                       double classValue,
                                                                       Instances workData)

Returns a dataset that contains all instances of a certain class value.

Parameters:

data - dataset to select the instances from

attrIndex - index of the relevant attribute

classIndex - index of the class attribute

classValue - the relevant class value

Returns:

a dataset with only

getInstancesFromClass

public static Instances getInstancesFromClass(Instances data,
                                              int classIndex,
                                              double classValue)

Returns a dataset that contains of all instances of a certain class value. Missing values are not dealt with.

Parameters:

data - dataset to select the instances from

classIndex - index of the class attribute

classValue - the class value

Returns:

a dataset with only instances of one class value

writeCurve

public static void writeCurve(java.lang.String f,
                              Estimator est,
                              double min,
                              double max,
                              int numPoints)
                       throws java.lang.Exception

Output of an n points of a density curve. Filename is parameter f + ".curv".

Parameters:

f - string to build filename

est -

min -

max -

numPoints -

Throws:

java.lang.Exception - if something goes wrong

writeCurve

public static void writeCurve(java.lang.String f,
                              Estimator est,
                              Estimator classEst,
                              double classIndex,
                              double min,
                              double max,
                              int numPoints)
                       throws java.lang.Exception

Output of an n points of a density curve. Filename is parameter f + ".curv".

Parameters:

f - string to build filename

est -

classEst -

classIndex -

min -

max -

numPoints -

Throws:

java.lang.Exception - if something goes wrong

getInstancesFromValue

public static Instances getInstancesFromValue(Instances data,
                                              int index,
                                              double v)

Returns a dataset that contains of all instances of a certain value for the given attribute. Does not deal with missing values.

Parameters:

data - dataset to select the instances from

index - the index of the attribute

v - the value

Returns:

a subdataset with only instances of one value for the attribute

cutpointsToString

public static java.lang.String cutpointsToString(double[] cutPoints,
                                                 boolean[] cutAndLeft)

Returns a string representing the cutpoints.

getRevision

public java.lang.String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Returns:

the revision