mulan.classifier.neural
Class DataPair

java.lang.Object
  mulan.classifier.neural.DataPair

public class DataPair
extends Object

Class for representation of a data-pair instance. The data pair contains an input pattern and respected true or expected output pattern for the input pattern.

Version:

2012.02.27

Author:

Jozef Vilcek

Constructor Summary

DataPair(double[] inputPattern, double[] trueOutput)
Creates a DataPair instance.

Method Summary

static List<DataPair>	createDataPairs(MultiLabelInstances mlDataSet, boolean bipolarOutput) Creates a DataPair representation for each Instance contained in MultiLabelInstances data set.
double[]	getInput() Gets the input pattern.
double[]	getOutput() Gets the ideal/expected output pattern.
boolean[]	getOutputBoolean() Gets the ideal/expected output pattern as boolean values.

Methods inherited from class java.lang.Object

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

Constructor Detail

DataPair

public DataPair(double[] inputPattern,
                double[] trueOutput)

Creates a DataPair instance.

Parameters:

inputPattern - the input pattern

trueOutput - the true/expected output pattern for the input

Method Detail

getInput

public double[] getInput()

Gets the input pattern.

Returns:

the input pattern

getOutput

public double[] getOutput()

Gets the ideal/expected output pattern.

Returns:

the output pattern

getOutputBoolean

public boolean[] getOutputBoolean()

Gets the ideal/expected output pattern as boolean values. This is useful when output represents labels bipartition. If output values in double[] are not in boolean representation, then output of this method is might not be valid. The computation is as follows:

- if value is equal to 1, then output is true in boolean

- if value is other than 1, then output is false in boolean

Returns:

the boolean representation of the output pattern

createDataPairs

public static List<DataPair> createDataPairs(MultiLabelInstances mlDataSet,
                                             boolean bipolarOutput)

Creates a DataPair representation for each Instance contained in MultiLabelInstances data set. The DataPair is a light weight representation of instance values (by double values), which is useful when iteration over the data and its values.

Parameters:

mlDataSet - the MultiLabelInstances which content has to be converted to list of DataPair

bipolarOutput - indicates whether output values should be converted to bipolar values, or left intact as binary

Returns:

the list of data pairs