org.neuroph.eval

Class FoldResult

java.lang.Object

org.neuroph.eval.FoldResult

public class FoldResult
extends Object

Result from single cross-validation fold, includes neural network, training and validation set, and fold evaluation results (at the moment only confsionMatrix) TODO: add different eveluation metrics, for regression too.

Author:

Nevena Mlenkovic

Constructor Summary

Constructors

Constructor and Description
FoldResult(NeuralNetwork neuralNet, DataSet trainingSet, DataSet validationSet)

Method Summary

Modifier and Type	Method and Description
ConfusionMatrix	getConfusionMatrix()
NeuralNetwork	getNeuralNet() Returns neural network trained in this cross-validation fold.
DataSet	getTrainingSet()
DataSet	getValidationSet()
void	setConfusionMatrix(ConfusionMatrix confusionMatrix)

Methods inherited from class java.lang.Object

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

Constructor Detail

FoldResult

public FoldResult(NeuralNetwork neuralNet,
                  DataSet trainingSet,
                  DataSet validationSet)

Method Detail

getNeuralNet

public NeuralNetwork getNeuralNet()

Returns neural network trained in this cross-validation fold.

Returns:

the neuralNet

getTrainingSet

public DataSet getTrainingSet()

Returns:

the trainingSet

getValidationSet

public DataSet getValidationSet()

Returns:

the validationSet

getConfusionMatrix

public ConfusionMatrix getConfusionMatrix()

Returns:

the confusionMatrix

setConfusionMatrix

public void setConfusionMatrix(ConfusionMatrix confusionMatrix)

Parameters:

confusionMatrix - the confusionMatrix to set