Serialized Form

Package org.neuroph.core

Class org.neuroph.core.Connection extends Object implements Serializable

serialVersionUID: 1L

Serialized Fields

fromNeuron

Neuron fromNeuron

From neuron for this connection (source neuron). This connection is output connection for from neuron.

toNeuron

Neuron toNeuron

To neuron for this connection (target, destination neuron) This connection is input connection for to neuron.

weight

Weight weight

Weight for this connection

Class org.neuroph.core.Layer extends Object implements Serializable

serialVersionUID: 3L

Serialized Fields

parentNetwork

NeuralNetwork parentNetwork

Reference to parent neural network

neurons

Neuron[] neurons

Array of neurons (Neuron instances)

label

String label

Label for this layer

Class org.neuroph.core.NeuralNetwork extends Observable implements Serializable

serialVersionUID: 4L

Serialized Fields

type

NeuralNetworkType type

Network type id (see neuroph.util.NeuralNetworkType)

layers

Layer[] layers

Neural network

output

double[] output

Neural network output buffer

inputNeurons

Neuron[] inputNeurons

Reference to network input neurons

outputNeurons

Neuron[] outputNeurons

Reference to newtwork output neurons

learningRule

LearningRule learningRule

Learning rule for this network

plugins

Map<K,V> plugins

Plugins collection

label

String label

Label for this network

Class org.neuroph.core.Neuron extends Object implements Serializable

serialVersionUID: 3L

Serialized Fields

parentLayer

Layer parentLayer

Parent layer for this neuron

inputConnections

Connection[] inputConnections

Array of neuron's input connections (connections to this neuron)

outConnections

Connection[] outConnections

Array of neuron's output connections (connections from this to other neurons)

inputFunction

InputFunction inputFunction

Input function for this neuron

transferFunction

TransferFunction transferFunction

Transfer function for this neuron

label

String label

Neuron's label

Class org.neuroph.core.Weight extends Object implements Serializable

serialVersionUID: 2L

Serialized Fields

value

double value

Weight value

Package org.neuroph.core.events

Class org.neuroph.core.events.LearningEvent extends EventObject implements Serializable

Package org.neuroph.core.exceptions

Class org.neuroph.core.exceptions.NeurophException extends RuntimeException implements Serializable

serialVersionUID: 0L

Class org.neuroph.core.exceptions.VectorSizeMismatchException extends NeurophException implements Serializable

serialVersionUID: 2L

Package org.neuroph.core.input

Class org.neuroph.core.input.And extends InputFunction implements Serializable

serialVersionUID: 1L

Class org.neuroph.core.input.Difference extends InputFunction implements Serializable

serialVersionUID: 21L

Class org.neuroph.core.input.InputFunction extends Object implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.Max extends InputFunction implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.Min extends InputFunction implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.Or extends InputFunction implements Serializable

serialVersionUID: 1L

Class org.neuroph.core.input.Product extends InputFunction implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.Sum extends InputFunction implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.SumSqr extends InputFunction implements Serializable

serialVersionUID: 2L

Class org.neuroph.core.input.WeightedSum extends InputFunction implements Serializable

serialVersionUID: 1L

Package org.neuroph.core.learning

Class org.neuroph.core.learning.DataSet extends Object implements Serializable

serialVersionUID: 2L

Serialized Fields

rows

List<E> rows

Collection of data rows

inputSize

int inputSize

outputSize

int outputSize

columnNames

String[] columnNames

label

String label

Label for this training set

Class org.neuroph.core.learning.DataSetRow extends Object implements Serializable

serialVersionUID: 1L

Serialized Fields

input

double[] input

Input vector for this training element

desiredOutput

double[] desiredOutput

Desired output for this training element

label

String label

Label for this training element

Class org.neuroph.core.learning.IterativeLearning extends LearningRule implements Serializable

serialVersionUID: 1L

Serialized Fields

learningRate

double learningRate

Learning rate parametar

currentIteration

int currentIteration

Current iteration counter

maxIterations

int maxIterations

Max training iterations (when to stopLearning training) TODO: this field should be private, to force use of setMaxIterations from derived classes, so iterationsLimited flag is also set at the sam etime.Wil that break backward compatibility with serialized networks?

iterationsLimited

boolean iterationsLimited

Flag for indicating if the training iteration number is limited

Class org.neuroph.core.learning.LearningRule extends Object implements Serializable

serialVersionUID: 1L

Serialized Fields

neuralNetwork

NeuralNetwork neuralNetwork

Neural network to train

listeners

EventListenerList listeners

List of learning rule listeners

Class org.neuroph.core.learning.SupervisedLearning extends IterativeLearning implements Serializable

serialVersionUID: 3L

Serialized Fields

maxError

double maxError

Max allowed network error (condition to stop learning)

batchMode

boolean batchMode

Setting to determine if learning (weights update) is in batch mode False by default.

outputError

double[] outputError

Stores network output error vector

trainingSetSize

int trainingSetSize

Class org.neuroph.core.learning.UnsupervisedLearning extends IterativeLearning implements Serializable

serialVersionUID: 1L

Package org.neuroph.core.transfer

Class org.neuroph.core.transfer.Gaussian extends TransferFunction implements Serializable

serialVersionUID: 1L

Serialized Fields

sigma

double sigma

The sigma parametetar of the gaussian function

Class org.neuroph.core.transfer.Linear extends TransferFunction implements Serializable

serialVersionUID: 1L

Serialized Fields

slope

double slope

The slope parametetar of the linear function

Class org.neuroph.core.transfer.Log extends TransferFunction implements Serializable

Class org.neuroph.core.transfer.Ramp extends TransferFunction implements Serializable

serialVersionUID: 1L

Serialized Fields

slope

double slope

The slope parametetar of the ramp function

xHigh

double xHigh

Threshold for the high output level

xLow

double xLow

Threshold for the low output level

yHigh

double yHigh

Output value for the high output level

yLow

double yLow

Output value for the low output level

Class org.neuroph.core.transfer.Sgn extends TransferFunction implements Serializable

serialVersionUID: 1L

Class org.neuroph.core.transfer.Sigmoid extends TransferFunction implements Serializable

serialVersionUID: 2L

Serialized Fields

slope

double slope

The slope parametetar of the sigmoid function

Class org.neuroph.core.transfer.Sin extends TransferFunction implements Serializable

Class org.neuroph.core.transfer.Step extends TransferFunction implements Serializable

serialVersionUID: 1L

Serialized Fields

yHigh

double yHigh

Output value for high output level

yLow

double yLow

Output value for low output level

Class org.neuroph.core.transfer.Tanh extends TransferFunction implements Serializable

serialVersionUID: 2L

Serialized Fields

slope

double slope

The slope parametetar of the Tanh function

Class org.neuroph.core.transfer.TransferFunction extends Object implements Serializable

serialVersionUID: 1L

Serialized Fields

output

double output

Output result

Class org.neuroph.core.transfer.Trapezoid extends TransferFunction implements Serializable

serialVersionUID: 1L

Serialized Fields

leftLow

double leftLow

leftHigh

double leftHigh

rightLow

double rightLow

rightHigh

double rightHigh

Package org.neuroph.nnet

Class org.neuroph.nnet.Adaline extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.BAM extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.CompetitiveNetwork extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.Hopfield extends NeuralNetwork implements Serializable

serialVersionUID: 2L

Class org.neuroph.nnet.Instar extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.Kohonen extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.MaxNet extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.MultiLayerPerceptron extends NeuralNetwork implements Serializable

serialVersionUID: 2L

Class org.neuroph.nnet.NeuroFuzzyPerceptron extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.Outstar extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.Perceptron extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.RbfNetwork extends NeuralNetwork implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.SupervisedHebbianNetwork extends NeuralNetwork implements Serializable

serialVersionUID: 2L

Class org.neuroph.nnet.UnsupervisedHebbianNetwork extends NeuralNetwork implements Serializable

serialVersionUID: 2L

Package org.neuroph.nnet.comp

Class org.neuroph.nnet.comp.DelayedConnection extends Connection implements Serializable

serialVersionUID: 1L

Serialized Fields

delay

int delay

Delay factor for this conection

Package org.neuroph.nnet.comp.layer

Class org.neuroph.nnet.comp.layer.CompetitiveLayer extends Layer implements Serializable

serialVersionUID: 1L

Serialized Fields

maxIterations

int maxIterations

Max iterations for neurons to compete This is neccesery to limit, otherwise if there is no winner there will be endless loop.

winner

CompetitiveNeuron winner

The competition winner for this layer

Class org.neuroph.nnet.comp.layer.InputLayer extends Layer implements Serializable

Package org.neuroph.nnet.comp.neuron

Class org.neuroph.nnet.comp.neuron.BiasNeuron extends Neuron implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.comp.neuron.CompetitiveNeuron extends DelayedNeuron implements Serializable

serialVersionUID: 1L

Serialized Fields

isCompeting

boolean isCompeting

Flag indicates if this neuron is in competing mode

connectionsFromOtherLayers

Connection[] connectionsFromOtherLayers

Collection of conections from neurons in other layers

connectionsFromThisLayer

Connection[] connectionsFromThisLayer

Collection of connections from neurons in the same layer as this neuron (lateral connections used for competition)

Class org.neuroph.nnet.comp.neuron.DelayedNeuron extends Neuron implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.comp.neuron.InputNeuron extends Neuron implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.comp.neuron.InputOutputNeuron extends Neuron implements Serializable

serialVersionUID: 1L

Serialized Fields

externalInputSet

boolean externalInputSet

Flag which is set true if neuron external input is set

bias

double bias

Bias value for this neuron

Class org.neuroph.nnet.comp.neuron.ThresholdNeuron extends Neuron implements Serializable

serialVersionUID: 1L

Serialized Fields

thresh

double thresh

Threshold value for this neuron

Package org.neuroph.nnet.learning

Class org.neuroph.nnet.learning.AntiHebbianLearning extends UnsupervisedHebbianLearning implements Serializable

Class org.neuroph.nnet.learning.BackPropagation extends SigmoidDeltaRule implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.BinaryDeltaRule extends PerceptronLearning implements Serializable

serialVersionUID: 1L

Serialized Fields

errorCorrection

double errorCorrection

The errorCorrection parametar of this learning algorithm

Class org.neuroph.nnet.learning.BinaryHebbianLearning extends UnsupervisedHebbianLearning implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.CompetitiveLearning extends UnsupervisedLearning implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.DynamicBackPropagation extends MomentumBackpropagation implements Serializable

serialVersionUID: 1L

Serialized Fields

maxLearningRate

double maxLearningRate

minLearningRate

double minLearningRate

learningRateChange

double learningRateChange

useDynamicLearningRate

boolean useDynamicLearningRate

maxMomentum

double maxMomentum

minMomentum

double minMomentum

momentumChange

double momentumChange

useDynamicMomentum

boolean useDynamicMomentum

Class org.neuroph.nnet.learning.GeneralizedHebbianLearning extends UnsupervisedHebbianLearning implements Serializable

Class org.neuroph.nnet.learning.HopfieldLearning extends LearningRule implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.InstarLearning extends UnsupervisedHebbianLearning implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.KohonenLearning extends LearningRule implements Serializable

serialVersionUID: 1L

Serialized Fields

learningRate

double learningRate

iterations

int[] iterations

decStep

double[] decStep

mapSize

int mapSize

nR

int[] nR

currentIteration

int currentIteration

Class org.neuroph.nnet.learning.LMS extends SupervisedLearning implements Serializable

serialVersionUID: 2L

Class org.neuroph.nnet.learning.MomentumBackpropagation extends BackPropagation implements Serializable

serialVersionUID: 1L

Serialized Fields

momentum

double momentum

Momentum factor

Class org.neuroph.nnet.learning.OjaLearning extends UnsupervisedHebbianLearning implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.OutstarLearning extends UnsupervisedHebbianLearning implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.PerceptronLearning extends LMS implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.ResilientPropagation extends BackPropagation implements Serializable

Serialized Fields

decreaseFactor

double decreaseFactor

increaseFactor

double increaseFactor

initialDelta

double initialDelta

maxDelta

double maxDelta

minDelta

double minDelta

Class org.neuroph.nnet.learning.SigmoidDeltaRule extends LMS implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.SimulatedAnnealingLearning extends SupervisedLearning implements Serializable

serialVersionUID: 1L

Serialized Fields

network

NeuralNetwork network

The neural network that is to be trained.

startTemperature

double startTemperature

The starting temperature.

stopTemperature

double stopTemperature

The ending temperature.

cycles

int cycles

The number of cycles that will be used.

temperature

double temperature

The current temperature.

weights

double[] weights

Current weights from the neural network.

bestWeights

double[] bestWeights

Best weights so far.

Class org.neuroph.nnet.learning.SupervisedHebbianLearning extends LMS implements Serializable

serialVersionUID: 1L

Class org.neuroph.nnet.learning.UnsupervisedHebbianLearning extends UnsupervisedLearning implements Serializable

serialVersionUID: 1L

Package org.neuroph.util

Class org.neuroph.util.NeuronProperties extends Properties implements Serializable

serialVersionUID: 2L

Class org.neuroph.util.Properties extends Hashtable implements Serializable

serialVersionUID: 1L

Package org.neuroph.util.io

Class org.neuroph.util.io.NeurophInputException extends NeurophException implements Serializable

serialVersionUID: 1L

Class org.neuroph.util.io.NeurophOutputException extends NeurophException implements Serializable

serialVersionUID: 1L

Package org.neuroph.util.plugins

Class org.neuroph.util.plugins.PluginBase extends Object implements Serializable

serialVersionUID: 1L

Serialized Fields

name

String name

Name for this plugin

parentNetwork

NeuralNetwork parentNetwork

Reference to parent neural network