C# Class numl.Supervised.NeuralNetwork.Neuron

An Artifical Neuron.

Inheritance: Data.IVertex

Protected Properties

Property	Type	Description
_DeltaL	double

Method	Description
Error ( double t, NetworkTrainingProperties properties ) : double	Calculates and returns the error derivative (Delta) of this node.
Evaluate ( ) : double	Calculates and returns the Node's Output value. Input is equal to the weights multiplied by the source Neuron's Output.
GetId ( ) : int	Gets the current identifier assigned when creating new nodes.
Neuron ( ) : System
Neuron ( bool isBias ) : System	Default constructor.
Reset ( NetworkTrainingProperties properties ) : void	Propogates a reset event downstream through the network.
SetId ( int startId ) : void	Sets the starting identifier used when creating new nodes. When rebuilding networks across sessions it is recommended that the identifier is set to the last used identifier.
ToString ( ) : string	Returns a string that represents the current object.
Update ( NetworkTrainingProperties properties ) : void	Propagates a weight update event upstream through the network using the supplied learning rate.

Calculates and returns the error derivative (Delta) of this node.

public Error ( double t, NetworkTrainingProperties properties ) : double
t	double	Error derivative from the previous layer (n + 1).
properties	NetworkTrainingProperties	Network training properties.
return	double

Calculates and returns the Node's Output value.

Input is equal to the weights multiplied by the source Neuron's Output.

public Evaluate ( ) : double
return	double

Gets the current identifier assigned when creating new nodes.

public static GetId ( ) : int
return	int

public Neuron ( ) : System
return	System

Default constructor.

public Neuron ( bool isBias ) : System
isBias	bool
return	System

Propogates a reset event downstream through the network.

public Reset ( NetworkTrainingProperties properties ) : void
properties	NetworkTrainingProperties	Network training properties.
return	void

Sets the starting identifier used when creating new nodes.

When rebuilding networks across sessions it is recommended that the identifier is set to the last used identifier.

public static SetId ( int startId ) : void
startId	int
return	void

Returns a string that represents the current object.

public ToString ( ) : string
return	string

Propagates a weight update event upstream through the network using the supplied learning rate.

public Update ( NetworkTrainingProperties properties ) : void
properties	NetworkTrainingProperties	Network training properties.
return	void

protected double _DeltaL
return	double