org.aika.network.neuron.lattice

Class InputNode

java.lang.Object

org.aika.network.neuron.lattice.Node

org.aika.network.neuron.lattice.InputNode

All Implemented Interfaces:

Comparable<Node>

Direct Known Subclasses:

NegativeInputNode, PositiveInputNode

public abstract class InputNode
extends Node

Author:

Lukas Molzberger

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	InputNode.Key

Field Summary

Fields

Modifier and Type	Field and Description
Integer	deactivationCount
Neuron	inputNeuron
InputNode.Key	key

Fields inherited from class org.aika.network.neuron.lattice.Node

activations, andChildren, andChildrenWithinDocument, countingMode, currentNodeId, frequency, frequencyHasChanged, id, isBlocked, isPredefined, isRemoved, isRemovedId, isRemovedIdCounter, level, minFrequency, n, neuron, orChildren

Constructor Summary

Constructors

Constructor and Description
InputNode(InputNode.Key key)

Method Summary

Modifier and Type	Method and Description
static InputNode	add(Neuron input, boolean isNeg, int posDelta, Integer deactivationCount)
abstract Activation	addActivation(Iteration t, Activation inputAct)
protected void	collectNodeAndRefinements(Set<InputNode> inputs)
double	computeSynapseWeightSum(Neuron n)
void	expandToNextLevel(Iteration t, LatticeQueue queue, Activation act, Option conflict, boolean train)
String	logicToString()
void	remove()
abstract void	removeActivation(Iteration t, Activation.Key inputAk)
void	setActivationsEmpty()

Methods inherited from class org.aika.network.neuron.lattice.Node

addActivationAndPropagate, addActivationInternal, addNeuron, cleanup, clearActivations, compareTo, computeAndParents, computePredefinedAndNodes, computePredefinedInputNodes, computePredefinedSecondLevelAndNodes, countActivation, getActivation, getActivation, getActivation, getActivation, getActivations, getActivations, getAndChildPatterns, getMatchingActivations, getWeight, isFrequentOrPredefined, isFrequentOrPredefined, propagateAddedActivation, propagateRemovedActivation, removeActivationAndPropagate, removeActivationInternal, removeFromNextLevel, toSimpleString, toString, train

Methods inherited from class java.lang.Object

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

Field Detail

inputNeuron

public Neuron inputNeuron

key

public InputNode.Key key

deactivationCount

public Integer deactivationCount

Constructor Detail

InputNode

public InputNode(InputNode.Key key)

Method Detail

addActivation

public abstract Activation addActivation(Iteration t,
                                         Activation inputAct)

removeActivation

public abstract void removeActivation(Iteration t,
                                      Activation.Key inputAk)

add

public static InputNode add(Neuron input,
                            boolean isNeg,
                            int posDelta,
                            Integer deactivationCount)

remove

public void remove()

Overrides:

remove in class Node

setActivationsEmpty

public void setActivationsEmpty()

Specified by:

setActivationsEmpty in class Node

collectNodeAndRefinements

protected void collectNodeAndRefinements(Set<InputNode> inputs)

Specified by:

collectNodeAndRefinements in class Node

expandToNextLevel

public void expandToNextLevel(Iteration t,
                              LatticeQueue queue,
                              Activation act,
                              Option conflict,
                              boolean train)

Specified by:

expandToNextLevel in class Node

computeSynapseWeightSum

public double computeSynapseWeightSum(Neuron n)

Specified by:

computeSynapseWeightSum in class Node

logicToString

public String logicToString()

Specified by:

logicToString in class Node