org.encog.neural.thermal
Class ThermalNetwork

java.lang.Object
  org.encog.ml.BasicML
      org.encog.neural.thermal.ThermalNetwork

All Implemented Interfaces:

Serializable, MLAutoAssocation, MLInput, MLInputOutput, MLMethod, MLOutput, MLProperties, MLRegression, MLResettable

Direct Known Subclasses:

BoltzmannMachine, HopfieldNetwork

public abstract class ThermalNetwork
extends BasicML
implements MLMethod, MLAutoAssocation, MLResettable

The thermal network forms the base class for Hopfield and Boltzmann machines.

Author:

jheaton

See Also:

Serialized Form

Constructor Summary

ThermalNetwork()
Default constructor.

ThermalNetwork(int neuronCount)
Construct the network with the specicified neuron count.

Method Summary

void	addWeight(int fromNeuron, int toNeuron, double value) Add to the specified weight.
double	calculateEnergy()
void	clear() Clear any connection weights.
BiPolarNeuralData	getCurrentState()
int	getNeuronCount()
double	getWeight(int fromNeuron, int toNeuron) Get a weight.
double[]	getWeights()
void	init(int neuronCount, double[] weights, double[] output) Init the network.
void	reset() Reset the weights.
void	reset(int seed) Reset the weights with a seed.
void	setCurrentState(BiPolarNeuralData state)
void	setCurrentState(double[] s) Set the current state.
void	setNeuronCount(int c) Set the neuron count.
void	setWeight(int fromNeuron, int toNeuron, double value) Set the weight.
void	setWeights(double[] w) Set the weight array.

Methods inherited from class org.encog.ml.BasicML

getProperties, getPropertyDouble, getPropertyLong, getPropertyString, setProperty, setProperty, setProperty, updateProperties

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.encog.ml.MLRegression

compute

Methods inherited from interface org.encog.ml.MLInput

getInputCount

Methods inherited from interface org.encog.ml.MLOutput

getOutputCount

Constructor Detail

ThermalNetwork

public ThermalNetwork()

Default constructor.

ThermalNetwork

public ThermalNetwork(int neuronCount)

Construct the network with the specicified neuron count.

Parameters:

neuronCount - The number of neurons.

Method Detail

addWeight

public final void addWeight(int fromNeuron,
                            int toNeuron,
                            double value)

Add to the specified weight.

Parameters:

fromNeuron - The from neuron.

toNeuron - The to neuron.

value - The value to add.

calculateEnergy

public final double calculateEnergy()

Returns:

Calculate the current energy for the network. The network will seek to lower this value.

clear

public final void clear()

Clear any connection weights.

getCurrentState

public final BiPolarNeuralData getCurrentState()

Returns:

The current state of the network.

getNeuronCount

public final int getNeuronCount()

Returns:

Get the neuron count for the network.

getWeight

public final double getWeight(int fromNeuron,
                              int toNeuron)

Get a weight.

Parameters:

fromNeuron - The from neuron.

toNeuron - The to neuron.

Returns:

The weight.

getWeights

public final double[] getWeights()

Returns:

The weights.

init

public final void init(int neuronCount,
                       double[] weights,
                       double[] output)

Init the network.

Parameters:

neuronCount - The neuron count.

weights - The weights.

output - The toutpu

reset

public final void reset()

Reset the weights.

Specified by:

reset in interface MLResettable

reset

public final void reset(int seed)

Reset the weights with a seed.

Specified by:

reset in interface MLResettable

Parameters:

seed - The seed value.

setCurrentState

public final void setCurrentState(BiPolarNeuralData state)

Parameters:

state - The current state for the network.

setCurrentState

public final void setCurrentState(double[] s)

Set the current state.

Parameters:

s - The current state array.

setNeuronCount

public final void setNeuronCount(int c)

Set the neuron count.

Parameters:

c - The neuron count.

setWeight

public final void setWeight(int fromNeuron,
                            int toNeuron,
                            double value)

Set the weight.

Parameters:

fromNeuron - The from neuron.

toNeuron - The to neuron.

value - The value.

setWeights

public final void setWeights(double[] w)

Set the weight array.

Parameters:

w - The weight array.