org.neuroph.core.input
Class InputFunction
java.lang.Object
org.neuroph.core.input.InputFunction
- All Implemented Interfaces:
- Serializable
- Direct Known Subclasses:
- And, Difference, 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
Method Summary |
abstract double |
getOutput(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 |
InputFunction
public InputFunction()
getOutput
public abstract double getOutput(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
Copyright © 2012. All Rights Reserved.