weka.filters.supervised.attribute
Class ClassOrder

java.lang.Object
  weka.filters.Filter
      weka.filters.supervised.attribute.ClassOrder

All Implemented Interfaces:

OptionHandler, java.io.Serializable, SupervisedFilter

public class ClassOrder

extends Filter

implements SupervisedFilter, OptionHandler

A filter that sorts the order of classes so that the class values are no longer of in the order of that in the header file after filtered. The values of the class will be in the order specified by the user -- it could be either in ascending/descending order by the class frequency or in random order.

The format of the header is thus not changed in this filter (although it still uses setInputFormat()), but the class value of each instance is converted to sorted values within the same range. The value can also be converted back using originalValue(double value) procedure.

Version:

$Revision: 1.4 $

Author:

Xin Xu (xx5@cs.waikato.ac.nz), Eibe Frank (eibe@cs.waikato.ac.nz)

See Also:

Serialized Form

Field Summary

static int	FREQ_ASCEND The class values are sorted in ascending order based on their frequencies
static int	FREQ_DESCEND The class values are sorted in descending order based on their frequencies
private Attribute	m_ClassAttribute Class attribute of the data
private double[]	m_ClassCounts This class can provide the class distribution in the sorted order as side effect
private int	m_ClassOrder The class order to be sorted
private int[]	m_Converter The 1-1 converting table from the original class values to the new values
private java.util.Random	m_Random The random object
private long	m_Seed The seed of randomization
static int	RANDOM The class values are sorted in random order

Fields inherited from class weka.filters.Filter

m_NewBatch

Constructor Summary

ClassOrder()

Method Summary

boolean	batchFinished() Signify that this batch of input to the filter is finished.
java.lang.String	classOrderTipText() Returns the tip text for this property
double[]	distributionsByOriginalIndex(double[] before) Convert the given class distribution back to the distributions with the original internal class index
double[]	getClassCounts() Get the class distribution of the sorted class values.
int	getClassOrder() Get the wanted class order
java.lang.String[]	getOptions() Gets the current settings of the filter.
long	getSeed() Get the current randomization seed
java.lang.String	globalInfo() Returns a string describing this filter
boolean	input(Instance instance) Input an instance for filtering.
java.util.Enumeration	listOptions() Returns an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class.
double	originalValue(double value) Return the original internal class value given the randomized class value, i.e. the string presentations of the two indices are the same.
java.lang.String	seedTipText() Returns the tip text for this property
void	setClassOrder(int order) Set the wanted class order
boolean	setInputFormat(Instances instanceInfo) Sets the format of the input instances.
void	setOptions(java.lang.String[] options) Parses a given list of options controlling the behaviour of this object.
void	setSeed(long seed) Set randomization seed

Methods inherited from class weka.filters.Filter

batchFilterFile, bufferInput, copyStringValues, copyStringValues, filterFile, flushInput, getInputFormat, getInputStringIndex, getOutputFormat, getOutputStringIndex, getStringIndices, inputFormat, inputFormatPeek, isOutputFormatDefined, numPendingOutput, output, outputFormat, outputFormatPeek, outputPeek, push, resetQueue, setOutputFormat, useFilter

Methods inherited from class java.lang.Object

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

Field Detail

m_Seed

private long m_Seed

The seed of randomization

m_Random

private java.util.Random m_Random

The random object

m_Converter

private int[] m_Converter

The 1-1 converting table from the original class values to the new values

m_ClassAttribute

private Attribute m_ClassAttribute

Class attribute of the data

m_ClassOrder

private int m_ClassOrder

The class order to be sorted

FREQ_ASCEND

public static final int FREQ_ASCEND

The class values are sorted in ascending order based on their frequencies

See Also:

Constant Field Values

FREQ_DESCEND

public static final int FREQ_DESCEND

The class values are sorted in descending order based on their frequencies

See Also:

Constant Field Values

RANDOM

public static final int RANDOM

The class values are sorted in random order

See Also:

Constant Field Values

m_ClassCounts

private double[] m_ClassCounts

This class can provide the class distribution in the sorted order as side effect

Constructor Detail

ClassOrder

public ClassOrder()

Method Detail

globalInfo

public java.lang.String globalInfo()

Returns a string describing this filter

Returns:

a description of the filter 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 controlling the behaviour of this object. Valid options are:

-R
Specify the seed of randomization used to randomize the class order (default: 1)

-C
Specify the class order to be sorted, could be 0: ascending, 1: descending and 2: random(default: 0)

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

getOptions

public java.lang.String[] getOptions()

Gets the current settings of the filter.

Specified by:

getOptions in interface OptionHandler

Returns:

an array of strings suitable for passing to setOptions

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

getSeed

public long getSeed()

Get the current randomization seed

Returns:

a seed

setSeed

public void setSeed(long seed)

Set randomization seed

Parameters:

seed - the set seed

classOrderTipText

public java.lang.String classOrderTipText()

Returns the tip text for this property

Returns:

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

getClassOrder

public int getClassOrder()

Get the wanted class order

Returns:

class order

setClassOrder

public void setClassOrder(int order)

Set the wanted class order

Parameters:

order - the class order

setInputFormat

public boolean setInputFormat(Instances instanceInfo)
                       throws java.lang.Exception

Sets the format of the input instances.

Overrides:

setInputFormat in class Filter

Parameters:

instanceInfo - an Instances object containing the input instance structure (any instances contained in the object are ignored - only the structure is required).

Returns:

true if the outputFormat may be collected immediately

Throws:

java.lang.Exception - if the inputFormat can't be set successfully

input

public boolean input(Instance instance)

Input an instance for filtering. Ordinarily the instance is processed and made available for output immediately. Some filters require all instances be read before producing output.

Overrides:

input in class Filter

Parameters:

instance - the input instance

Returns:

true if the filtered instance may now be collected with output().

Throws:

java.lang.IllegalStateException - if no input format has been defined.

batchFinished

public boolean batchFinished()
                      throws java.lang.Exception

Signify that this batch of input to the filter is finished. If the filter requires all instances prior to filtering, output() may now be called to retrieve the filtered instances. Any subsequent instances filtered should be filtered based on setting obtained from the first batch (unless the inputFormat has been re-assigned or new options have been set). This implementation sorts the class values and provide class counts in the output format

Overrides:

batchFinished in class Filter

Returns:

true if there are instances pending output

Throws:

java.lang.NullPointerException - if no input structure has been defined,

java.lang.Exception - if there was a problem finishing the batch.

getClassCounts

public double[] getClassCounts()

Get the class distribution of the sorted class values. If class is numeric it returns null

Returns:

the class counts

distributionsByOriginalIndex

public double[] distributionsByOriginalIndex(double[] before)

Convert the given class distribution back to the distributions with the original internal class index

Parameters:

before - the given class distribution

Returns:

the distribution converted back

originalValue

public double originalValue(double value)
                     throws java.lang.Exception

Return the original internal class value given the randomized class value, i.e. the string presentations of the two indices are the same. It's useful when the filter is used within a classifier so that the filtering procedure should be transparent to the evaluation

Parameters:

value - the given value

Returns:

the original internal value, -1 if not found

Throws:

if - the coverter table is not set yet

java.lang.Exception

main

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

Main method for testing this class.

Parameters:

argv - should contain arguments to the filter: use -h for help