javax.datamining.algorithm.feedforwardneuralnet
Interface FeedForwardNeuralNetSettings

All Superinterfaces:

AlgorithmSettings, SupervisedAlgorithmSettings

public interface FeedForwardNeuralNetSettings

extends SupervisedAlgorithmSettings

A FeedForwardNeuralNetSettings object captures the parameters associated with a neural network algorithm. It allows a knowledgeable user to fine tune algorithm parameters. Generally, not all parameters must be specified, however, those specified are taken into account by the underlying data mining engine.

Author:

JSR-73 Java Data Mining Expert Group

See Also:

AlgorithmSettings, SupervisedAlgorithmSettings

Method Summary

boolean	determineNumberOfNodesPerLayer() Returns true if the number of nodes is determined by the DME.
void	determineNumberOfNodesPerLayer(boolean determineNumberOfNodesPerLayer) The number of nodes is determined by the DME, if true.
LearningAlgorithm	getLearningAlgorithm() Returns the LearningAlgorithm object to be used for model building.
int	getMaxNumberOfIterations() Returns the maximum number of iterations to be used in training the network.
double	getMinErrorTolerance() Returns the minimum acceptable error in modeling the training data.
NeuralLayer[]	getNeuralLayers() Returns the set of layers associated with a model.
void	setLearningAlgorithm(LearningAlgorithm learningAlgorithm) Sets the learning algorithm type to be used for model building.
void	setMaxNumberOfIterations(int maxIterations) Sets the maximum number of iterations to be used in training the network.
void	setMinErrorTolerance(double minTolerance) Sets the minimum acceptable error in modeling the training data.
void	setNeuralLayers(NeuralLayer[] hiddenLayerArray) Adds the array of hidden layers to the settings, starting at the input side.

Methods inherited from interface javax.datamining.base.AlgorithmSettings

getMiningAlgorithm, verify

Method Detail

determineNumberOfNodesPerLayer

public boolean determineNumberOfNodesPerLayer()

Returns true if the number of nodes is determined by the DME. Returns false otherwise. Initially, this method returns the vendor default value.

Returns:

boolean

determineNumberOfNodesPerLayer

public void determineNumberOfNodesPerLayer(boolean determineNumberOfNodesPerLayer)

The number of nodes is determined by the DME, if true. If false, the number is determined by the user for each layer.

Parameters:

determineNumberOfNodesPerLayer -

Returns:

void

getLearningAlgorithm

public LearningAlgorithm getLearningAlgorithm()

Returns the LearningAlgorithm object to be used for model building.

Returns:

LearningAlgorithm

getMaxNumberOfIterations

public int getMaxNumberOfIterations()

Returns the maximum number of iterations to be used in training the network.

Returns:

int

getMinErrorTolerance

public double getMinErrorTolerance()

Returns the minimum acceptable error in modeling the training data.

Returns:

double

getNeuralLayers

public NeuralLayer[] getNeuralLayers()

Returns the set of layers associated with a model.

Returns:

NeuralLayer[]

setLearningAlgorithm

public void setLearningAlgorithm(LearningAlgorithm learningAlgorithm)

Sets the learning algorithm type to be used for model building.

Parameters:

learningAlgorithm - The learning algorithm to be used to build a feed forward neural network model.

Returns:

void

setMaxNumberOfIterations

public void setMaxNumberOfIterations(int maxIterations)

Sets the maximum number of iterations to be used in training the network. The number of iterations must be a positive integer.

Parameters:

maxIterations - The maximum number of iterations to be used in training the network.

Returns:

void

setMinErrorTolerance

public void setMinErrorTolerance(double minTolerance)

Sets the minimum acceptable error in modeling the training data. The minimum tolerance must be a non-negative number that is less than 1.

Parameters:

minTolerance - The minimum acceptable error in modeling the training data.

Returns:

void

setNeuralLayers

public void setNeuralLayers(NeuralLayer[] hiddenLayerArray)

Adds the array of hidden layers to the settings, starting at the input side.

Parameters:

hiddenLayerArray - An array of neural layers to be added.

Returns:

void