org.neuroph.nnet.comp
Class CompetitiveNeuron

java.lang.Object
  extended by org.neuroph.core.Neuron
      extended by org.neuroph.nnet.comp.DelayedNeuron
          extended by org.neuroph.nnet.comp.CompetitiveNeuron
All Implemented Interfaces:
java.io.Serializable

public class CompetitiveNeuron
extends DelayedNeuron

Provides neuron behaviour specific for competitive neurons which are used in competitive layers, and networks with competitive learning.

Author:
Zoran Sevarac
See Also:
Serialized Form

Field Summary
 
Fields inherited from class org.neuroph.nnet.comp.DelayedNeuron
outputHistory
 
Fields inherited from class org.neuroph.core.Neuron
error, inputConnections, inputFunction, netInput, outConnections, output, parentLayer, transferFunction
 
Constructor Summary
CompetitiveNeuron(InputFunction inputFunction, TransferFunction transferFunction)
          Creates an instance of CompetitiveNeuron with specified input and transfer functions
 
Method Summary
 void addInputConnection(Connection connection)
          Adds input connection for this competitive neuron
 void calculate()
          Calculates neuron's output
 java.util.Vector<Connection> getConnectionsFromOtherLayers()
          Returns collection of connections from other layers
 boolean isCompeting()
          Retruns true if this neuron is in competing mode, false otherwise
 void reset()
          Resets the input, output and mode for this neuron
 void setIsCompeting(boolean isCompeting)
          Sets the flag to indicate that this neuron is in competing mode
 
Methods inherited from class org.neuroph.nnet.comp.DelayedNeuron
getOutput
 
Methods inherited from class org.neuroph.core.Neuron
addInputConnection, addInputConnection, addOutputConnection, getConnectionFrom, getError, getInputConnections, getInputFunction, getInputsIterator, getNetInput, getOutConnections, getOutput, getParentLayer, getTransferFunction, getWeightsVector, hasInputConnections, randomizeInputWeights, removeInputConnectionFrom, setError, setInput, setInputFunction, setOutput, setParentLayer, setTransferFunction
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

CompetitiveNeuron

public CompetitiveNeuron(InputFunction inputFunction,
                         TransferFunction transferFunction)
Creates an instance of CompetitiveNeuron with specified input and transfer functions

Parameters:
inputFunction - neuron input function
transferFunction - neuron ransfer function
Method Detail

calculate

public void calculate()
Description copied from class: Neuron
Calculates neuron's output

Overrides:
calculate in class DelayedNeuron

addInputConnection

public void addInputConnection(Connection connection)
Adds input connection for this competitive neuron

Overrides:
addInputConnection in class Neuron
Parameters:
connection - input connection

getConnectionsFromOtherLayers

public java.util.Vector<Connection> getConnectionsFromOtherLayers()
Returns collection of connections from other layers

Returns:
collection of connections from other layers

reset

public void reset()
Resets the input, output and mode for this neuron

Overrides:
reset in class Neuron

isCompeting

public boolean isCompeting()
Retruns true if this neuron is in competing mode, false otherwise

Returns:
true if this neuron is in competing mode, false otherwise

setIsCompeting

public void setIsCompeting(boolean isCompeting)
Sets the flag to indicate that this neuron is in competing mode

Parameters:
isCompeting - value for the isCompeting flag