org.encog.engine.data
Class BasicEngineData

java.lang.Object
  org.encog.engine.data.BasicEngineData

All Implemented Interfaces:

Serializable, EngineData

public class BasicEngineData
extends Object
implements EngineData, Serializable

A basic implementation of the EngineData interface. This implementation simply holds and input and ideal NeuralData object. For supervised training both input and ideal should be specified. For unsupervised training the input property should be valid, however the ideal property should contain null.

Author:

jheaton

See Also:

Serialized Form

Constructor Summary

BasicEngineData(double[] input)
Construct the object with only input.

BasicEngineData(double[] input, double[] ideal)
Construct a BasicNeuralDataPair class with the specified input and ideal values.

Method Summary

static EngineData	createPair(int inputSize, int idealSize) Create a new neural data pair object of the correct size for the neural network that is being trained.
double[]	getIdealArray() Get the expected results.
double[]	getInputArray() Get the input data.
boolean	isSupervised() Determine if this data pair is supervised.
void	setIdealArray(double[] data) Set the ideal array.
void	setInputArray(double[] data) Set the input array.
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

BasicEngineData

public BasicEngineData(double[] input)

Construct the object with only input. If this constructor is used, then unsupervised training is being used.

Parameters:

input - The input to the neural network.

BasicEngineData

public BasicEngineData(double[] input,
                       double[] ideal)

Construct a BasicNeuralDataPair class with the specified input and ideal values.

Parameters:

input - The input to the neural network.

ideal - The expected results from the neural network.

Method Detail

createPair

public static EngineData createPair(int inputSize,
                                    int idealSize)

Create a new neural data pair object of the correct size for the neural network that is being trained. This object will be passed to the getPair method to allow the neural data pair objects to be copied to it.

Parameters:

inputSize - The size of the input data.

idealSize - The size of the ideal data.

Returns:

A new neural data pair object.

getIdealArray

public double[] getIdealArray()

Get the expected results. Returns null if this is unsupervised training.

Specified by:

getIdealArray in interface EngineData

Returns:

Returns the expected results, or null if unsupervised training.

getInputArray

public double[] getInputArray()

Get the input data.

Specified by:

getInputArray in interface EngineData

Returns:

The input data.

isSupervised

public boolean isSupervised()

Determine if this data pair is supervised.

Specified by:

isSupervised in interface EngineData

Returns:

True if this data pair is supervised.

setIdealArray

public void setIdealArray(double[] data)

Set the ideal array.

Specified by:

setIdealArray in interface EngineData

Parameters:

data - The ideal array.

setInputArray

public void setInputArray(double[] data)

Set the input array.

Specified by:

setInputArray in interface EngineData

Parameters:

data - The input array.

toString

public String toString()

Overrides:

toString in class Object