java.lang.Object
- network.aika.AbstractNode<Neuron>
- - network.aika.neuron.INeuron

All Implemented Interfaces:

Comparable<INeuron>, Writable
```
public class INeuron
extends AbstractNode<Neuron>
implements Comparable<INeuron>
```
The INeuron class represents a internal neuron implementation in Aikas neural network and is connected to other neurons through input synapses and output synapses. The activation value of a neuron is calculated by computing the weighted sum (input act. value * synapse weight) of the input synapses, adding the bias to it and sending the resulting value through a transfer function (the upper part of tanh).

The neuron does not store its activationsBySlotAndPosition by itself. The activation objects are stored within the logic nodes. To access the activationsBySlotAndPosition of this neuron simply use the member variable node or use the method getFinalActivations(Document doc) to ge the final activationsBySlotAndPosition of this neuron.

Author:

Lukas Molzberger

Nested Class Summary

Nested Classes
Modifier and Type Class Description

static class INeuron.SynapseSummary

static class INeuron.Type

Field Summary

Fields
Modifier and Type Field Description

static INeuron MAX_NEURON

static INeuron MIN_NEURON

static double WEIGHT_TOLERANCE
- Fields inherited from class network.aika.AbstractNode
  lastUsedDocumentId, modified, provider

Constructor Summary

Constructors
Constructor	Description
`INeuron(Model m, String label, String outputText, INeuron.Type type, ActivationFunction actF)`
`INeuron(Model m, String label, INeuron.Type type, ActivationFunction actF)`
`INeuron(Neuron p)`

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method	Description
`Activation`	`addInput(Document doc, Activation.Builder input)`	Propagate an input activation into the network.
`void`	`addModelLabel(String modelLabel)`
`void`	`clearActivations()`
`void`	`clearActivations(int threadId)`
`void`	`clearActivations(Document doc)`
`void`	`commit(Collection<Synapse> modifiedSynapses)`
`int`	`compareTo(INeuron n)`
`protected Activation`	`createActivation(Document doc, Map<Integer,Position> slots)`
`void`	`delete(Set<String> modelLabels)`
`ActivationFunction`	`getActivationFunction()`
`Stream<Activation>`	`getActivations(Document doc)`
`Stream<Activation>`	`getActivations(Document doc, boolean onlyFinal)`
`Stream<Activation>`	`getActivations(Document doc, int slot, Position pos, boolean onlyFinal)`
`Stream<Activation>`	`getActivations(Document doc, int fromSlot, Position fromPos, boolean fromInclusive, int toSlot, Position toPos, boolean toInclusive)`
`Collection<Activation>`	`getActivations(Document doc, SortedMap<Integer,Position> slots)`
`double`	`getBias()`
`double`	`getBiasDelta()`
`Integer`	`getId()`
`Provider<OrNode>`	`getInputNode()`
`Collection<Synapse>`	`getInputSynapses()`
`String`	`getLabel()`
`Synapse`	`getMaxInputSynapse(Synapse.State state)`
`Model`	`getModel()`
`double`	`getNewBias()`
`int`	`getNewSynapseId()`
`Provider<InputNode>`	`getOutputNode()`
`Map<Integer,Relation>`	`getOutputRelations()`
`Collection<Synapse>`	`getOutputSynapses()`
`String`	`getOutputText()`
`Collection<Synapse>`	`getPassiveInputSynapses()`
`INeuron.SynapseSummary`	`getSynapseSummary()`
`double`	`getTotalBias(Synapse.State state)`
`INeuron.Type`	`getType()`
`boolean`	`isEmpty(Document doc)`
`boolean`	`isNeuron()`
`boolean`	`isPassiveInputNeuron()`
`void`	`propagate(Activation act)`
`void`	`reactivate()`
`void`	`readFields(DataInput in, Model m)`	Deserialize the fields of this object from `in`.
`void`	`register(Activation act)`
`void`	`registerPassiveInputSynapse(Synapse s)`
`void`	`registerSynapseId(Integer synId)`
`void`	`remove()`
`void`	`setActivationFunction(ActivationFunction actF)`
`void`	`setBias(double b)`
`void`	`setOutputNode(Provider<InputNode> node)`
`void`	`setOutputText(String outputText)`
`void`	`setType(INeuron.Type t)`
`int`	`size(Document doc)`
`void`	`suspend()`
`protected String`	`toDetailedString()`
`String`	`toString()`
`String`	`toStringWithSynapses()`
`String`	`typeToString()`
`void`	`updateBiasDelta(double biasDelta)`
`void`	`write(DataOutput out)`	Serialize the fields of this object to `out`.

Methods inherited from class network.aika.AbstractNode
getModelLabels, getProvider, read, setModified

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

WEIGHT_TOLERANCE
```
public static double WEIGHT_TOLERANCE
```

MIN_NEURON
```
public static final INeuron MIN_NEURON
```

MAX_NEURON
```
public static final INeuron MAX_NEURON
```

Constructor Detail

INeuron
```
public INeuron(Neuron p)
```

INeuron

public INeuron(Model m,
               String label,
               INeuron.Type type,
               ActivationFunction actF)

INeuron

public INeuron(Model m,
               String label,
               String outputText,
               INeuron.Type type,
               ActivationFunction actF)

Method Detail

setOutputNode

public void setOutputNode(Provider<InputNode> node)

getId
```
public Integer getId()
```

getLabel
```
public String getLabel()
```
Overrides:

getLabel in class AbstractNode<Neuron>

getType
```
public INeuron.Type getType()
```

setType

public void setType(INeuron.Type t)

getOutputNode

public Provider<InputNode> getOutputNode()

getInputNode

public Provider<OrNode> getInputNode()

getSynapseSummary

public INeuron.SynapseSummary getSynapseSummary()

getOutputRelations

public Map<Integer,Relation> getOutputRelations()

getInputSynapses

public Collection<Synapse> getInputSynapses()

addModelLabel
```
public void addModelLabel(String modelLabel)
```
Overrides:

addModelLabel in class AbstractNode<Neuron>

getMaxInputSynapse

public Synapse getMaxInputSynapse(Synapse.State state)

getOutputSynapses

public Collection<Synapse> getOutputSynapses()

getPassiveInputSynapses

public Collection<Synapse> getPassiveInputSynapses()

getActivationFunction

public ActivationFunction getActivationFunction()

setActivationFunction

public void setActivationFunction(ActivationFunction actF)

getActivations

public Stream<Activation> getActivations(Document doc)

isEmpty

public boolean isEmpty(Document doc)

size
```
public int size(Document doc)
```

clearActivations

public void clearActivations(Document doc)

getActivations

public Stream<Activation> getActivations(Document doc,
                                         int slot,
                                         Position pos,
                                         boolean onlyFinal)

clearActivations
```
public void clearActivations()
```

clearActivations

public void clearActivations(int threadId)

getModel
```
public Model getModel()
```

getActivations

public Stream<Activation> getActivations(Document doc,
                                         int fromSlot,
                                         Position fromPos,
                                         boolean fromInclusive,
                                         int toSlot,
                                         Position toPos,
                                         boolean toInclusive)

getActivations

public Stream<Activation> getActivations(Document doc,
                                         boolean onlyFinal)

getActivations

public Collection<Activation> getActivations(Document doc,
                                             SortedMap<Integer,Position> slots)

setOutputText

public void setOutputText(String outputText)

getOutputText
```
public String getOutputText()
```

addInput

public Activation addInput(Document doc,
                           Activation.Builder input)

Propagate an input activation into the network.

Parameters:: doc - The current document; input -

createActivation

protected Activation createActivation(Document doc,
                                      Map<Integer,Position> slots)

getTotalBias

public double getTotalBias(Synapse.State state)

commit

public void commit(Collection<Synapse> modifiedSynapses)

remove
```
public void remove()
```

getNewSynapseId
```
public int getNewSynapseId()
```

registerSynapseId

public void registerSynapseId(Integer synId)

propagate

public void propagate(Activation act)

compareTo
```
public int compareTo(INeuron n)
```
Specified by:

compareTo in interface Comparable<INeuron>

write
```
public void write(DataOutput out)
           throws IOException
```
Description copied from interface: Writable

Serialize the fields of this object to out.

Specified by:

write in interface Writable

Overrides:

write in class AbstractNode<Neuron>

Parameters:

out - DataOuput to serialize this object into.

Throws:

IOException

readFields
```
public void readFields(DataInput in,
                       Model m)
                throws IOException
```
Description copied from interface: Writable

Deserialize the fields of this object from in.
For efficiency, implementations should attempt to re-use storage in the existing object where possible.

Specified by:

readFields in interface Writable

Overrides:

readFields in class AbstractNode<Neuron>

Parameters:

in - DataInput to deseriablize this object from.

Throws:

IOException

suspend
```
public void suspend()
```
Overrides:

suspend in class AbstractNode<Neuron>

reactivate
```
public void reactivate()
```
Overrides:

reactivate in class AbstractNode<Neuron>

delete
```
public void delete(Set<String> modelLabels)
```
Specified by:

delete in class AbstractNode<Neuron>

isNeuron
```
public boolean isNeuron()
```
Specified by:

isNeuron in class AbstractNode<Neuron>

setBias
```
public void setBias(double b)
```

updateBiasDelta

public void updateBiasDelta(double biasDelta)

getBias
```
public double getBias()
```

getNewBias
```
public double getNewBias()
```

getBiasDelta
```
public double getBiasDelta()
```

register

public void register(Activation act)

isPassiveInputNeuron
```
public boolean isPassiveInputNeuron()
```

registerPassiveInputSynapse

public void registerPassiveInputSynapse(Synapse s)

toString
```
public String toString()
```
Overrides:

toString in class Object

toDetailedString
```
protected String toDetailedString()
```

typeToString
```
public String typeToString()
```

toStringWithSynapses
```
public String toStringWithSynapses()
```

Modifier and Type	Class	Description
`static class`	`INeuron.SynapseSummary`
`static class`	`INeuron.Type`

Modifier and Type	Field	Description
`static INeuron`	`MAX_NEURON`
`static INeuron`	`MIN_NEURON`
`static double`	`WEIGHT_TOLERANCE`

Class INeuron

Nested Class Summary

Field Summary

Fields inherited from class network.aika.AbstractNode

Constructor Summary

Method Summary

Methods inherited from class network.aika.AbstractNode

Methods inherited from class java.lang.Object

Field Detail

WEIGHT_TOLERANCE

MIN_NEURON

MAX_NEURON

Constructor Detail

INeuron

INeuron

INeuron

Method Detail

setOutputNode

getId

getLabel

getType

setType

getOutputNode

getInputNode

getSynapseSummary

getOutputRelations

getInputSynapses

addModelLabel

getMaxInputSynapse

getOutputSynapses

getPassiveInputSynapses

getActivationFunction

setActivationFunction

getActivations

isEmpty

size

clearActivations

getActivations

clearActivations

clearActivations

getModel

getActivations

getActivations

getActivations

setOutputText

getOutputText

addInput

createActivation

getTotalBias

commit

remove

getNewSynapseId

registerSynapseId

propagate

compareTo

write

readFields

suspend

reactivate

delete

isNeuron

setBias

updateBiasDelta

getBias

getNewBias

getBiasDelta

register

isPassiveInputNeuron

registerPassiveInputSynapse

toString

toDetailedString

typeToString

toStringWithSynapses