marf.Classification.NeuralNetwork
Class Neuron

java.lang.Object
  java.lang.Thread
      marf.util.BaseThread
          marf.Classification.NeuralNetwork.Neuron

All Implemented Interfaces:

java.io.Serializable, java.lang.Runnable

public class Neuron
extends BaseThread
implements java.io.Serializable

Class Neuron -- a basic element of a neural network.

$Id: Neuron.java,v 1.19 2007/12/31 00:17:04 mokhov Exp $

Since:

0.0.1

Version:

$Revision: 1.19 $

Author:

Ian Clement, Serguei Mokhov

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class java.lang.Thread

java.lang.Thread.State, java.lang.Thread.UncaughtExceptionHandler

Field Summary

protected double	dDelta Used in error calculation.
protected double	dResult Current Neuron's result.
protected double	dThreshold Activation threshold.
static int	HIDDEN Indicates middle (hidden) neuron.
static int	INPUT Indicates input neuron.
protected int	iType Current neuron type.
static int	OUTPUT Indicates output neuron.
protected java.lang.String	strName Neuron's name.
static int	UNDEF Indicates undefined neuron type.

Fields inherited from class marf.util.BaseThread

iTID, oTarget, siNextTID

Fields inherited from class java.lang.Thread

MAX_PRIORITY, MIN_PRIORITY, NORM_PRIORITY

Constructor Summary

Neuron(java.lang.String pstrName, int piType)
Neuron's Constructor.

Method Summary

boolean	addInput(Neuron poInputNeuron, double pdWeight) Adds an input neuron and its associated weight.
boolean	addOutput(Neuron poOutputNeuron) Adds an output neuron.
void	commit() Applies weight changes.
void	eval() Evaluates current neuron's value.
static java.lang.String	getMARFSourceCodeRevision() Retrieves class' revision.
void	printXML(java.io.BufferedWriter poWriter, int piTabsNum) Dumps XML of the current Neuron.
void	run() The body of a Neuron thread.
void	train(double pdExpected, double pdAlpha, double pdBeta) Neuron training.

Methods inherited from class marf.util.BaseThread

getNextTID, getTarget, getTID

Methods inherited from class java.lang.Thread

activeCount, checkAccess, countStackFrames, currentThread, destroy, dumpStack, enumerate, getAllStackTraces, getContextClassLoader, getDefaultUncaughtExceptionHandler, getId, getName, getPriority, getStackTrace, getState, getThreadGroup, getUncaughtExceptionHandler, holdsLock, interrupt, interrupted, isAlive, isDaemon, isInterrupted, join, join, join, resume, setContextClassLoader, setDaemon, setDefaultUncaughtExceptionHandler, setName, setPriority, setUncaughtExceptionHandler, sleep, sleep, start, stop, stop, suspend, toString, yield

Methods inherited from class java.lang.Object

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

Field Detail

UNDEF

public static final int UNDEF

Indicates undefined neuron type.

Since:

0.1.2

See Also:

Constant Field Values

INPUT

public static final int INPUT

Indicates input neuron.

Since:

0.1.2

See Also:

Constant Field Values

HIDDEN

public static final int HIDDEN

Indicates middle (hidden) neuron.

Since:

0.1.2

See Also:

Constant Field Values

OUTPUT

public static final int OUTPUT

Indicates output neuron.

Since:

0.1.2

See Also:

Constant Field Values

strName

protected java.lang.String strName

Neuron's name.

iType

protected int iType

Current neuron type.

dDelta

protected double dDelta

Used in error calculation.

dThreshold

protected double dThreshold

Activation threshold.

dResult

protected double dResult

Current Neuron's result.

Constructor Detail

Neuron

public Neuron(java.lang.String pstrName,
              int piType)

Neuron's Constructor.

Parameters:

pstrName - Neuron's name

piType - Neuron's type

Method Detail

addInput

public final boolean addInput(Neuron poInputNeuron,
                              double pdWeight)

Adds an input neuron and its associated weight.

Parameters:

poInputNeuron - Neuron to be added

pdWeight - associated weight

Returns:

true if add was successful

addOutput

public final boolean addOutput(Neuron poOutputNeuron)

Adds an output neuron.

Parameters:

poOutputNeuron - the Neuron to add

Returns:

true if the neuron was added

eval

public final void eval()

Evaluates current neuron's value.

train

public final void train(double pdExpected,
                        double pdAlpha,
                        double pdBeta)

Neuron training.

Parameters:

pdExpected - expected value

pdAlpha - used in error calculation (training constant)

pdBeta - used in error calculation

commit

public final void commit()

Applies weight changes.

printXML

public final void printXML(java.io.BufferedWriter poWriter,
                           int piTabsNum)
                    throws java.io.IOException

Dumps XML of the current Neuron.

Parameters:

poWriter - Writer object to write output to

piTabsNum - tabulation needed

Throws:

java.io.IOException - if there is any problem to output to the writer

run

public void run()

The body of a Neuron thread. TODO: complete.

Specified by:

run in interface java.lang.Runnable

Overrides:

run in class java.lang.Thread

Since:

0.3.0.2

getMARFSourceCodeRevision

public static java.lang.String getMARFSourceCodeRevision()

Retrieves class' revision.

Returns:

revision string

Since:

0.3.0.2