org.encog.neural.networks.training.cross

Class NetworkFold

java.lang.Object

org.encog.neural.networks.training.cross.NetworkFold

public class NetworkFold
extends java.lang.Object

The network for one fold of a cross validation.

Constructor Summary

Constructors

Constructor and Description
NetworkFold(FlatNetwork flat) Construct a fold from the specified flat network.

Method Summary

Modifier and Type	Method and Description
void	copyFromNetwork(FlatNetwork source) Copy the weights and output from the network.
void	copyToNetwork(FlatNetwork target) Copy weights and output to the network.
double[]	getOutput()
double[]	getWeights()

Methods inherited from class java.lang.Object

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

Constructor Detail

NetworkFold

public NetworkFold(FlatNetwork flat)

Construct a fold from the specified flat network.

Parameters:

flat - THe flat network.

Method Detail

copyToNetwork

public final void copyToNetwork(FlatNetwork target)

Copy weights and output to the network.

Parameters:

target - The network to copy to.

copyFromNetwork

public final void copyFromNetwork(FlatNetwork source)

Copy the weights and output from the network.

Parameters:

source - The network to copy from.

getWeights

public final double[] getWeights()

Returns:

The network weights.

getOutput

public final double[] getOutput()

Returns:

The network output.