org.neuroph.core.input

Class InputFunction

java.lang.Object

org.neuroph.core.input.InputFunction

All Implemented Interfaces:

Serializable

Direct Known Subclasses:

And, Difference, EuclideanRBF, Max, Min, Or, Product, Sum, SumSqr, WeightedSum

public abstract class InputFunction
extends Object
implements Serializable

 Neuron's input function. It has two subcomponents:
 
 weightsFunction - which performs operation with input and weight vector
 summingFunction - which performs operation with the resulting vector from weightsFunction
 
 InputFunction implements the following behaviour:
 output = summingFunction(weightsFunction(inputs))
 
 Different neuron input functions can be created by setting different weights and summing functions.

Author:

Zoran Sevarac

See Also:

Neuron, Serialized Form

Constructor Summary

Constructors

Constructor and Description
InputFunction()

Method Summary

Modifier and Type	Method and Description
abstract double	getOutput(List<Connection> inputConnections) Returns ouput value of this input function for the given neuron inputs

Methods inherited from class java.lang.Object

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

Constructor Detail

InputFunction

public InputFunction()

Method Detail

getOutput

public abstract double getOutput(List<Connection> inputConnections)

Returns ouput value of this input function for the given neuron inputs

Parameters:

inputConnections - neuron's input connections

Returns:

input total net input