org.simbrain.network.subnetworks

Class Hopfield

java.lang.Object

org.simbrain.network.groups.Group

org.simbrain.network.groups.Subnetwork

org.simbrain.network.subnetworks.Hopfield

All Implemented Interfaces:

Trainable

public class Hopfield
extends Subnetwork
implements Trainable

Hopfield is a basic implementation of a discrete Hopfield network.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Hopfield.HopfieldUpdate Main forms of Hopfield update rule.

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_NUM_UNITS Default number of neurons.
static Hopfield.HopfieldUpdate	DEFAULT_UPDATE Default update mechanism.

Constructor Summary

Constructors

Constructor and Description
Hopfield(Network root, int numNeurons) Creates a new Hopfield network.

Method Summary

Modifier and Type	Method and Description
static double	bipolar(double in) Convenience method to convert binary values (1,0) to bipolar values(1,-1).
java.util.List<Neuron>	getInputNeurons() Returns the list of input neurons.
Group	getNetwork() Returns a reference to the subnetwork or synapse group being trained.
java.util.HashSet<Neuron>	getNeuronSet()
int	getNumUnits()
java.util.List<Neuron>	getOutputNeurons() Returns the list of output neurons.
TrainingSet	getTrainingSet() Returns the the training set, which contains input and target data.
Hopfield.HopfieldUpdate	getUpdateFunc()
java.lang.String	getUpdateMethodDesecription() Returns a description of this group's update method, which is displayed in the update manager panel.
void	initNetwork() Initialize the network.
boolean	isByPriority()
void	randomize() Randomize weights symmetrically.
void	randomizeSequence() Randomizes the update sequence by shuffling the neuron list associated with this Hopfield network.
void	setByPriority(boolean byPriority)
void	setNeuronSet(java.util.HashSet<Neuron> neuronSet)
void	setUpdateFunc(Hopfield.HopfieldUpdate updateFunc)
void	trainOnCurrentPattern() Apply the basic Hopfield rule to the current pattern.
void	update() Update this group.

Methods inherited from class org.simbrain.network.groups.Subnetwork

addAndLabelSynapseGroup, addNeuronGroup, addRowToTrainingSet, addSynapseGroup, clearActivations, connectNeuronGroups, connectNeuronGroups, connectNeuronGroups, delete, displayNeuronGroups, getEnabled, getFlatNeuronList, getFlatSynapseList, getIndexOfNeuronGroup, getLongDescription, getModifiableNeuronList, getNeuronGroup, getNeuronGroup, getNeuronGroupByLabel, getNeuronGroupCount, getNeuronGroupList, getNeuronGroupsAsList, getSynapseGroup, getSynapseGroup, getSynapseGroupByLabel, getSynapseGroupCount, getSynapseGroupList, isEmpty, recursivelySetIds, removeNeuronGroup, removeSynapseGroup, setDisplayNeuronGroups, setEnabled, size, toString

Methods inherited from class org.simbrain.network.groups.Group

getId, getLabel, getParentGroup, getParentNetwork, getStateInfo, hasParentGroup, isMarkedForDeletion, isTopLevelGroup, setId, setLabel, setMarkedForDeletion, setParentGroup, setStateInfo

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_UPDATE

public static final Hopfield.HopfieldUpdate DEFAULT_UPDATE

Default update mechanism.

DEFAULT_NUM_UNITS

public static final int DEFAULT_NUM_UNITS

Default number of neurons.

See Also:

Constant Field Values

Constructor Detail

Hopfield

public Hopfield(Network root,
                int numNeurons)

Creates a new Hopfield network.

Parameters:

numNeurons - Number of neurons in new network

root - reference to Network.

Method Detail

randomizeSequence

public void randomizeSequence()

Randomizes the update sequence by shuffling the neuron list associated with this Hopfield network. Only has an effect if the update function is sequential.

randomize

public void randomize()

Randomize weights symmetrically.

update

public void update()

Description copied from class: Group

Update this group.

Overrides:

update in class Subnetwork

getNumUnits

public int getNumUnits()

Returns:

The number of neurons.

getUpdateMethodDesecription

public java.lang.String getUpdateMethodDesecription()

Description copied from class: Group

Returns a description of this group's update method, which is displayed in the update manager panel.

Overrides:

getUpdateMethodDesecription in class Subnetwork

Returns:

a description of the update method.

getInputNeurons

public java.util.List<Neuron> getInputNeurons()

Description copied from interface: Trainable

Returns the list of input neurons.

Specified by:

getInputNeurons in interface Trainable

Returns:

the list of input neurons.

getOutputNeurons

public java.util.List<Neuron> getOutputNeurons()

Description copied from interface: Trainable

Returns the list of output neurons.

Specified by:

getOutputNeurons in interface Trainable

Returns:

list of output neurons

getTrainingSet

public TrainingSet getTrainingSet()

Description copied from interface: Trainable

Returns the the training set, which contains input and target data.

Specified by:

getTrainingSet in interface Trainable

Returns:

the training set.

initNetwork

public void initNetwork()

Description copied from interface: Trainable

Initialize the network. E.g. used before a training run for SRN's to clear activations.

Specified by:

initNetwork in interface Trainable

trainOnCurrentPattern

public void trainOnCurrentPattern()

Apply the basic Hopfield rule to the current pattern. This is not the main training algorithm, which directly makes use of the input data.

bipolar

public static double bipolar(double in)

Convenience method to convert binary values (1,0) to bipolar values(1,-1).

Parameters:

in - number to convert

Returns:

converted number

getUpdateFunc

public Hopfield.HopfieldUpdate getUpdateFunc()

Returns:

the updateFunc

setUpdateFunc

public void setUpdateFunc(Hopfield.HopfieldUpdate updateFunc)

Parameters:

updateFunc - the updateFunc to set

getNeuronSet

public java.util.HashSet<Neuron> getNeuronSet()

Returns:

the neuronSet

setNeuronSet

public void setNeuronSet(java.util.HashSet<Neuron> neuronSet)

Parameters:

neuronSet - the neuronSet to set

isByPriority

public boolean isByPriority()

Returns:

the byPriority

setByPriority

public void setByPriority(boolean byPriority)

Parameters:

byPriority - the byPriority to set

getNetwork

public Group getNetwork()

Description copied from interface: Trainable

Returns a reference to the subnetwork or synapse group being trained.

Specified by:

getNetwork in interface Trainable

Returns:

the network being trained.