org.cognity.data

Class DataExample

java.lang.Object

org.cognity.data.DataExample

public class DataExample
extends Object

A data example is a basic data structure used during neural network training. The data example consists of an input vector and a target output vector. For unsupervised patterns the target vector can be omitted.

Author:

Artur Markiewicz

Constructor Summary

Constructors

Constructor and Description
DataExample(double[] input) Creates an unsupervised data example with specified input vector.
DataExample(double[] input, double[] target) Creates a supervised data example with specified input and target vector.

Method Summary

Methods

Modifier and Type	Method and Description
boolean	equals(Object obj) Two data examples are equal if and only if input and target vectors are equal.
double[]	getInput() Returns a reference to the input vector.
int	getInputSize() Returns the size of the input vector.
double[]	getTarget() Returns a reference to the target output vector.
int	getTargetSize() Returns the size of the target output vector.
int	hashCode()
boolean	isSupervised() Returns true if this data example is supervised.

Methods inherited from class java.lang.Object

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

Constructor Detail

DataExample

public DataExample(double[] input)

Creates an unsupervised data example with specified input vector. The array is copied to the object, so later changes in the input array, don't alter the data example.

Parameters:

input - the input vector. Cannot be null.

DataExample

public DataExample(double[] input,
           double[] target)

Creates a supervised data example with specified input and target vector. The array is copied to the object, so later changes in the input and target arrays, don't alter the data example.

Parameters:

input - the input vector. Cannot be null.

target - the target vector. Cannot be null.

Method Detail

getInput

public final double[] getInput()

Returns a reference to the input vector. Changes in the obtained array WILL be reflected in this object.

Returns:

a reference to the input vector

getTarget

public final double[] getTarget()

Returns a reference to the target output vector. Changes in the obtained array WILL be reflected in this object. If the data example is unsupervised, an array of size zero is returned.

Returns:

• a reference to the target output vector - for supervised examples
• a reference to an empty array - for unsupervised examples

getInputSize

public final int getInputSize()

Returns the size of the input vector.

Returns:

the size of the input vector

getTargetSize

public final int getTargetSize()

Returns the size of the target output vector.

Returns:

• the size of the target output vector - for supervised examples
• zero - for unsupervised examples

isSupervised

public final boolean isSupervised()

Returns true if this data example is supervised. Otherwise it returns false.

Returns:

true if this data example is supervised

equals

public final boolean equals(Object obj)

Two data examples are equal if and only if input and target vectors are equal.

Overrides:

equals in class Object

hashCode

public final int hashCode()

Overrides:

hashCode in class Object