keel.Algorithms.Neural_Networks.ensemble

Class EnsembleNetwork

java.lang.Object

keel.Algorithms.Neural_Networks.net.Network

keel.Algorithms.Neural_Networks.ensemble.EnsembleNetwork

public class EnsembleNetwork
extends Network

Class representing an ensemble of neural networks

Since:

JDK1.5

Version:

0.1

Author:

Written by Nicolas Garcia Pedrajas (University of Cordoba) 27/02/2007

Field Summary

Fields

Modifier and Type	Field and Description
Sample	sample Training data sample

Fields inherited from class keel.Algorithms.Neural_Networks.net.Network

a, activation, b_htan, b_log, delta, momentum, Nhidden, Ninputs, Nlayers, Noutputs, transfer, w

Constructor Summary

Constructors

Constructor and Description
EnsembleNetwork() Empty constructor
EnsembleNetwork(EnsembleParameters global) Constructor.

Method Summary

Modifier and Type	Method and Description
void	TrainNetwork(EnsembleParameters global, double[][] data, int npatterns) Back-propagation algorithm without cross validation
void	TrainNetworkWithCrossvalidation(EnsembleParameters global, Data data) Back-propagation algorithm using cross validation

Methods inherited from class keel.Algorithms.Neural_Networks.net.Network

GenerateOutput, GenerateOutput, LoadNetwork, NetClassifyPattern, NetGetClassOfPattern, PrintWeights, SaveNetwork, SaveOutputFile, TestNetworkInClassification, TestNetworkInRegression, TrainNetwork, TrainNetworkWithCrossvalidation

Methods inherited from class java.lang.Object

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

Field Detail

sample

public Sample sample

Training data sample

Constructor Detail

EnsembleNetwork

public EnsembleNetwork()

Empty constructor

EnsembleNetwork

public EnsembleNetwork(EnsembleParameters global)

Constructor.

Parameters:

global - Global definition parameters

Method Detail

TrainNetworkWithCrossvalidation

public void TrainNetworkWithCrossvalidation(EnsembleParameters global,
                                            Data data)

Back-propagation algorithm using cross validation

Parameters:

global - Global definition parameters

data - Input data

TrainNetwork

public void TrainNetwork(EnsembleParameters global,
                         double[][] data,
                         int npatterns)

Back-propagation algorithm without cross validation

Parameters:

global - Global definition parameters

data - Input data

npatterns - No of patterns