org.neuroph.contrib.model.modelselection

Class MultilayerPerceptronOptimazer<T extends BackPropagation>

java.lang.Object

org.neuroph.contrib.model.modelselection.MultilayerPerceptronOptimazer<T>

Type Parameters:

T - Type which defined which LearningRule will be used during model optimization

All Implemented Interfaces:

NeurophModelOptimizer

public class MultilayerPerceptronOptimazer<T extends BackPropagation>
extends java.lang.Object
implements NeurophModelOptimizer

Constructor Summary

Constructors

Constructor and Description
MultilayerPerceptronOptimazer() If ErrorEstimationMethod is not provided use KFoldCrossValidation by default

Method Summary

Modifier and Type	Method and Description
NeuralNetwork	createOptimalModel(DataSet dataSet)
MultilayerPerceptronOptimazer	withErrorEstimationMethod(CrossValidation errorEstimationMethod)
MultilayerPerceptronOptimazer	withLearningRule(BackPropagation learningRule)
MultilayerPerceptronOptimazer	withMaxLayers(int maxLayers)
MultilayerPerceptronOptimazer	withMaxNeurons(int maxNeurons)
MultilayerPerceptronOptimazer	withMinNeurons(int minNeurons)
MultilayerPerceptronOptimazer	withNeuronIncrement(int neuronIncrement)

Methods inherited from class java.lang.Object

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

Constructor Detail

MultilayerPerceptronOptimazer

public MultilayerPerceptronOptimazer()

If ErrorEstimationMethod is not provided use KFoldCrossValidation by default

Method Detail

withMaxLayers

public MultilayerPerceptronOptimazer withMaxLayers(int maxLayers)

withNeuronIncrement

public MultilayerPerceptronOptimazer withNeuronIncrement(int neuronIncrement)

withMaxNeurons

public MultilayerPerceptronOptimazer withMaxNeurons(int maxNeurons)

withMinNeurons

public MultilayerPerceptronOptimazer withMinNeurons(int minNeurons)

withErrorEstimationMethod

public MultilayerPerceptronOptimazer withErrorEstimationMethod(CrossValidation errorEstimationMethod)

withLearningRule

public MultilayerPerceptronOptimazer withLearningRule(BackPropagation learningRule)

createOptimalModel

public NeuralNetwork createOptimalModel(DataSet dataSet)

Specified by:

createOptimalModel in interface NeurophModelOptimizer

Parameters:

dataSet - training set used for error estimation

Returns:

neural network model with optimized architecture for provided data set