org.neuroph.core.transfer

Class TransferFunction

java.lang.Object

org.neuroph.core.transfer.TransferFunction

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

Gaussian, Linear, Log, Ramp, RectifiedLinear, Sgn, Sigmoid, Sin, Step, Tanh, Trapezoid

public abstract class TransferFunction
extends Object
implements Serializable

Abstract base class for all neuron tranfer functions.

Author:

Zoran Sevarac

See Also:

Neuron, Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected double	output Output result

Constructor Summary

Constructors

Constructor and Description
TransferFunction()

Method Summary

Modifier and Type	Method and Description
double	getDerivative(double totalInput) Returns the first derivative of this function.
abstract double	getOutput(double totalInput) Returns the ouput of this function.

Methods inherited from class java.lang.Object

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

Field Detail

output

protected transient double output

Output result

Constructor Detail

TransferFunction

public TransferFunction()

Method Detail

getOutput

public abstract double getOutput(double totalInput)

Returns the ouput of this function.

Parameters:

totalInput - total input

getDerivative

public double getDerivative(double totalInput)

Returns the first derivative of this function. Note: should this method should be abstract?

Parameters:

totalInput - total input