weka.filters
Class SplitDatasetFilter

java.lang.Object
  |
  +--weka.filters.Filter
        |
        +--weka.filters.SplitDatasetFilter

All Implemented Interfaces:

OptionHandler, java.io.Serializable

public class SplitDatasetFilter

extends Filter

implements OptionHandler

This filter takes a dataset and outputs a subset of it. If a class attribute is assigned, the dataset will be stratified when fold-based splitting. Valid options are:

-R inst1,inst2-inst4,...
Specifies list of instances to select. First and last are valid indexes. (default fold-based splitting)

-V
Specifies if inverse of selection is to be output.

-N number of folds
Specifies number of folds dataset is split into (default 10).

-F fold
Specifies which fold is selected. (default 1)

-S seed
Specifies a random number seed for shuffling the dataset. (default 0, don't randomize)

-A
If set, data is not being stratified even if class index is set.

Author:

Eibe Frank (eibe@cs.waikato.ac.nz)

See Also:

Serialized Form

Fields inherited from class weka.filters.Filter

m_NewBatch

Constructor Summary

SplitDatasetFilter()

Method Summary

boolean	batchFinished() Signify that this batch of input to the filter is finished.
boolean	getDontStratifyData() Gets whether stratification is not performed.
int	getFold() Gets the fold which is selected.
java.lang.String	getInstancesIndices() Gets ranges of instances selected.
boolean	getInvertSelection() Gets if selection is to be inverted.
int	getNumFolds() Gets the number of folds in which dataset is to be split into.
java.lang.String[]	getOptions() Gets the current settings of the filter.
long	getSeed() Gets the random number seed used for shuffling the dataset.
java.util.Enumeration	listOptions() Gets an enumeration describing the available options.
static void	main(java.lang.String[] argv) Main method for testing this class.
void	setDontStratifyData(boolean flag) Sets whether stratification is not performed.
void	setFold(int fold) Selects a fold.
boolean	setInputFormat(Instances instanceInfo) Sets the format of the input instances.
void	setInstancesIndices(java.lang.String rangeList) Sets the ranges of instances to be selected.
void	setInvertSelection(boolean inverse) Sets if selection is to be inverted.
void	setNumFolds(int numFolds) Sets the number of folds the dataset is split into.
void	setOptions(java.lang.String[] options) Parses the options for this object.
void	setSeed(long seed) Sets the random number seed for shuffling the dataset.

Methods inherited from class weka.filters.Filter

batchFilterFile, bufferInput, copyStringValues, copyStringValues, filterFile, flushInput, getInputFormat, getInputStringIndex, getOutputFormat, getOutputStringIndex, getStringIndices, input, inputFormat, 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

Constructor Detail

SplitDatasetFilter

public SplitDatasetFilter()

Method Detail

listOptions

public java.util.Enumeration listOptions()

Gets 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 the options for this object. Valid options are:

-R inst1,inst2-inst4,...
Specifies list of instances to select. First and last are valid indexes. (default fold-based splitting)

-V
Specifies if inverse of selection is to be output.

-N number of folds
Specifies number of folds dataset is split into (default 10).

-F fold
Specifies which fold is selected. (default 1)

-S seed
Specifies a random number seed for shuffling the dataset. (default 0, no randomizing)

-A
If set, data is not being stratified even if class index is set.

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

getInstancesIndices

public java.lang.String getInstancesIndices()

Gets ranges of instances selected.

Returns:

a string containing a comma-separated list of ranges

setInstancesIndices

public void setInstancesIndices(java.lang.String rangeList)

Sets the ranges of instances to be selected. If provided string is null, ranges won't be used for selecting instances.

Parameters:

rangeList - a string representing the list of instances. eg: first-3,5,6-last

Throws:

java.lang.IllegalArgumentException - if an invalid range list is supplied

getInvertSelection

public boolean getInvertSelection()

Gets if selection is to be inverted.

Returns:

true if the selection is to be inverted

setInvertSelection

public void setInvertSelection(boolean inverse)

Sets if selection is to be inverted.

Parameters:

inverse - true if inversion is to be performed

getNumFolds

public int getNumFolds()

Gets the number of folds in which dataset is to be split into.

Returns:

the number of folds the dataset is to be split into.

setNumFolds

public void setNumFolds(int numFolds)

Sets the number of folds the dataset is split into. If the number of folds is zero, it won't split it into folds.

Parameters:

numFolds - number of folds dataset is to be split into

Throws:

java.lang.IllegalArgumentException - if number of folds is negative

getFold

public int getFold()

Gets the fold which is selected.

Returns:

the fold which is selected

setFold

public void setFold(int fold)

Selects a fold.

Parameters:

fold - the fold to be selected.

Throws:

java.lang.IllegalArgumentException - if fold's index is smaller than 1

getSeed

public long getSeed()

Gets the random number seed used for shuffling the dataset.

Returns:

the random number seed

setSeed

public void setSeed(long seed)

Sets the random number seed for shuffling the dataset. If seed is negative, shuffling won't be performed.

Parameters:

seed - the random number seed

setDontStratifyData

public void setDontStratifyData(boolean flag)

Sets whether stratification is not performed.

getDontStratifyData

public boolean getDontStratifyData()

Gets whether stratification is not performed.

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 because outputFormat can be collected immediately

Throws:

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

batchFinished

public boolean batchFinished()

Signify that this batch of input to the filter is finished. Output() may now be called to retrieve the filtered instances.

Overrides:

batchFinished in class Filter

Returns:

true if there are instances pending output

Throws:

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

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