edu.udo.cs.yale.tools
Class WekaInstancesAdaptor

java.lang.Object
  weka.core.Instances
      edu.udo.cs.yale.tools.WekaInstancesAdaptor

All Implemented Interfaces:

java.io.Serializable

public class WekaInstancesAdaptor
extends weka.core.Instances

This class extends the Weka class Instances and overrides all methods needed to directly use a Yale ExampleSet as source for Weka instead of copying the complete data.

Version:

$Id: WekaInstancesAdaptor.java,v 2.13 2006/04/06 15:23:38 ingomierswa Exp $

Author:

Ingo Mierswa

See Also:

Serialized Form

Nested Class Summary

private class

WekaInstancesAdaptor.InstanceEnumeration This enumeration implementation uses an ExampleReader (Iterator) to enumerate the instances.

Field Summary

static int	ASSOCIATION_RULE_MINING
static int	CLUSTERING
private ExampleSet	exampleSet The example set which backs up the Instances object.
private FastExample2SparseTransform	exampleTransform This transformation might help to speed up the creation of sparse examples.
private Attribute	labelAttribute The YALE label attribute or null if not desired (depending on task).
static int	LEARNING
private int[]	mostFrequent The most frequent nominal values (only used for association rule mining, null otherwise)
static int	PREDICTING
private static long	serialVersionUID
private int	taskType The task type for which this instances object is used.
static int	WEIGHTING

Fields inherited from class weka.core.Instances

FILE_EXTENSION, m_Attributes, m_ClassIndex, m_IndicesBuffer, m_Instances, m_RelationName, m_ValueBuffer, SERIALIZED_OBJ_FILE_EXTENSION

Constructor Summary

WekaInstancesAdaptor(java.lang.String name, ExampleSet exampleSet, int taskType)
Creates a new Instances object based on the given example set.

Method Summary

java.util.Enumeration	enumerateInstances() Returns an instance enumeration based on an ExampleReader.
private static weka.core.FastVector	getAttributeVector(ExampleSet exampleSet, int taskType)
weka.core.Instance	instance(int i) Returns the i-th instance.
int	numInstances() Returns the number of instances.
private static weka.core.Attribute	toWekaAttribute(Attribute attribute) Converts an Attribute to a Weka attribute.
private weka.core.Instance	toWekaInstance(Example example) Gets an example and creates a Weka instance.

Methods inherited from class weka.core.Instances

add, attribute, attribute, attributeStats, attributeToDoubleArray, checkForAttributeType, checkForStringAttributes, checkInstance, classAttribute, classIndex, compactify, copyInstances, delete, delete, deleteAttributeAt, deleteAttributeType, deleteStringAttributes, deleteWithMissing, deleteWithMissing, deleteWithMissingClass, enumerateAttributes, equalHeaders, errms, firstInstance, freshAttributeInfo, getFirstToken, getIndex, getInstance, getInstanceFull, getInstanceSparse, getLastToken, getNextToken, getRandomNumberGenerator, initTokenizer, insertAttributeAt, instancesAndWeights, kthSmallestValue, kthSmallestValue, lastInstance, main, meanOrMode, meanOrMode, mergeInstances, numAttributes, numClasses, numDistinctValues, numDistinctValues, parseAttribute, partition, quickSort, randomize, readHeader, readInstance, readTillEOL, relationName, renameAttribute, renameAttribute, renameAttributeValue, renameAttributeValue, resample, resampleWithWeights, resampleWithWeights, select, setClass, setClassIndex, setRelationName, sort, sort, stratify, stratStep, stringFreeStructure, stringWithoutHeader, sumOfWeights, swap, test, testCV, toString, toSummaryString, trainCV, trainCV, variance, variance

Methods inherited from class java.lang.Object

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

Field Detail

serialVersionUID

private static final long serialVersionUID

See Also:

Constant Field Values

LEARNING

public static final int LEARNING

See Also:

Constant Field Values

PREDICTING

public static final int PREDICTING

See Also:

Constant Field Values

CLUSTERING

public static final int CLUSTERING

See Also:

Constant Field Values

ASSOCIATION_RULE_MINING

public static final int ASSOCIATION_RULE_MINING

See Also:

Constant Field Values

WEIGHTING

public static final int WEIGHTING

See Also:

Constant Field Values

exampleSet

private ExampleSet exampleSet

The example set which backs up the Instances object.

exampleTransform

private FastExample2SparseTransform exampleTransform

This transformation might help to speed up the creation of sparse examples.

mostFrequent

private int[] mostFrequent

The most frequent nominal values (only used for association rule mining, null otherwise). -1 if attribute is numerical.

taskType

private int taskType

The task type for which this instances object is used. Must be one out of LEARNING, PREDICTING, CLUSTERING, ASSOCIATION_RULE_MINING, or WEIGHTING. For the latter cases the original label attribute will be omitted.

labelAttribute

private Attribute labelAttribute

The YALE label attribute or null if not desired (depending on task).

Constructor Detail

WekaInstancesAdaptor

public WekaInstancesAdaptor(java.lang.String name,
                            ExampleSet exampleSet,
                            int taskType)

Creates a new Instances object based on the given example set.

Method Detail

enumerateInstances

public java.util.Enumeration enumerateInstances()

Returns an instance enumeration based on an ExampleReader.

Overrides:

enumerateInstances in class weka.core.Instances

instance

public weka.core.Instance instance(int i)

Returns the i-th instance.

Overrides:

instance in class weka.core.Instances

numInstances

public int numInstances()

Returns the number of instances.

Overrides:

numInstances in class weka.core.Instances

toWekaInstance

private weka.core.Instance toWekaInstance(Example example)

Gets an example and creates a Weka instance.

getAttributeVector

private static weka.core.FastVector getAttributeVector(ExampleSet exampleSet,
                                                       int taskType)

toWekaAttribute

private static weka.core.Attribute toWekaAttribute(Attribute attribute)

Converts an Attribute to a Weka attribute.