GeneralActivationFunction Class Reference
[Neural Network filtering]

The base class for all activation functions in neural network. More...

#include <GeneralActivationFunction.h>

Inheritance diagram for GeneralActivationFunction:

Inheritance graph
[legend]

List of all members.

Public Member Functions

virtual double CalcOutput (const double input)=0
 The function that maps between the weighted sum that's the input of the neuron and the output.
virtual double CalcDeriv (const double input)=0
 The derivative of the function defined in CalcOutput.
 GeneralActivationFunction ()
virtual ~GeneralActivationFunction ()


Detailed Description

The base class for all activation functions in neural network.

The activation function in a neural network node determines the type of non-linearity between input and output. We have to implement the activation function and its derivative.

Definition at line 13 of file GeneralActivationFunction.h.


Constructor & Destructor Documentation

GeneralActivationFunction::GeneralActivationFunction (  ) 

Definition at line 3 of file GeneralActivationFunction.cpp.

GeneralActivationFunction::~GeneralActivationFunction (  )  [virtual]

Definition at line 5 of file GeneralActivationFunction.cpp.


Member Function Documentation

virtual double GeneralActivationFunction::CalcOutput ( const double  input  )  [pure virtual]

The function that maps between the weighted sum that's the input of the neuron and the output.

Implemented in BipolarActivationFunction, and IdentityActivationFunction.

virtual double GeneralActivationFunction::CalcDeriv ( const double  input  )  [pure virtual]

The derivative of the function defined in CalcOutput.

Implemented in BipolarActivationFunction, and IdentityActivationFunction.


The documentation for this class was generated from the following files:

Generated on Fri Jul 4 15:30:21 2008 for GPLIB++ by  doxygen 1.5.5