javax.datamining.algorithm.feedforwardneuralnet
Interface NeuralLayer

public interface NeuralLayer

A NeuralLayer object captures the parameters required to describe a layer in a neural network model.

The vendor documentation specifies the default number of nodes, if used.

Author:

JSR-73 Java Data Mining Expert Group

Method Summary

ActivationFunction	getActivationFunction() Returns the ActivationFunction associated with the layer.
int	getNumberOfNodes() Returns the number of nodes in the layer.
void	setActivationFunction(ActivationFunction function) Sets the activation function for the layer.
void	setNumberOfNodes(int nodes) Sets the number of nodes in the layer.
boolean	useBias() Returns true if a bias term is used in the layer.
void	useBias(boolean useBias) Specifies whether a bias term is to be used in the layer.

Method Detail

getActivationFunction

public ActivationFunction getActivationFunction()

Returns the ActivationFunction associated with the layer. An activation function computes the activation state of a neuron based on its inputs.

Returns:

ActivationFunction

getNumberOfNodes

public int getNumberOfNodes()

Returns the number of nodes in the layer.

Returns:

int

setActivationFunction

public void setActivationFunction(ActivationFunction function)

Sets the activation function for the layer.

Parameters:

function - The activation function to be used in the layer.

Returns:

void

setNumberOfNodes

public void setNumberOfNodes(int nodes)

Sets the number of nodes in the layer.

Parameters:

nodes - The number of neurons in the layer.

Returns:

void

useBias

public boolean useBias()

Returns true if a bias term is used in the layer. A bias is equivalent to an input connection set at a constant level. Initially, this method returns the vendor default value.

Returns:

boolean

useBias

public void useBias(boolean useBias)

Specifies whether a bias term is to be used in the layer.

Parameters:

useBias - Indicator to use bias in the layer, if true.

Returns:

void