A B C D G H I M O R S T W X

A

adjustWeight() - Method in class Synapse
Adjusts the weight

B

BiasNeuron - class BiasNeuron.
The BiasNeuron class creates Bias Neurons which have a fixed value of -1
BiasNeuron() - Constructor for class BiasNeuron
This constructor for the BiasNeuroninitialises it's input value to -1

C

ContextNeuron - class ContextNeuron.
The ContextNeuron class creates Context Neuron which are used to provide recurrency in the RNN network.
ContextNeuron(double) - Constructor for class ContextNeuron
This constructor for the ContextNeuron specifies the memory depth and sets the input value to 0.5
ContextSynapse - class ContextSynapse.
The ContextSynapse class is used to connect Context Neurons to Hidden Neurons.
ContextSynapse(Neuron, Neuron) - Constructor for class ContextSynapse
This constructor for the ContextSynapse connects two Neurons together, and sets it's weight to 1
calculateDelayedOutput() - Method in class HiddenNeuron
Calculates the summation of the delayed output values
calculateDelayedOutput() - Method in class InputNeuron
Calculates the summation of the delayed output values
calculateHiddenError() - Method in class TDNN
Calculates the hidden error term
calculateHiddenOutput() - Method in class TDNN
Calculates the output of the Hidden Neurons
calculateHiddenWeightChange() - Method in class TDNN
Calculates the weight change to be made to the Synapses between the Hidden and Input Neurons and Bias Neurons
calculateOutput() - Method in class BiasNeuron
Calculates the output value
calculateOutput() - Method in class ContextNeuron
Calculates the output value
calculateOutput() - Method in class InputNeuron
Calculates the output value
calculateOutputWeightChange() - Method in class TDNN
Calculates the weight change to be made to the Synapses between the Output and Hidden Neurons and Bias Neurons
calculateWeightChange(double, double, double) - Method in class Synapse
Calculates the weight change
calculateWeightChange(double, double, double, int) - Method in class Synapse
Calculates the average weight change for use with time delays
closeFile() - Method in class OutputFile
Closes the file
connectNeurons() - Method in class RNN
Connects the Neurons
createNeurons(double) - Method in class RNN
Creates the Neurons
createNeurons(int) - Method in class TDNN
Creates the Neurons

D

DataProcessor - class DataProcessor.
The DataProcessor class is used to scale the input data between 0 and 1
DataProcessor(double[]) - Constructor for class DataProcessor
This constructor for the DataProcessor sets it's inputs

G

getData(String, String) - Method in class Test
Gets the data from the Document
getDocument() - Method in class XMLParser
Gets the Document
getOutputs() - Method in class HiddenNeuron
gets the delayed output values
getOutputs() - Method in class InputNeuron
gets the delayed output values
getWeight() - Method in class Synapse
Gets the Synapse weight

H

HiddenNeuron - class HiddenNeuron.
The HiddenNeuron class is used to create Hidden Neurons
HiddenNeuron(int, int) - Constructor for class HiddenNeuron
This constructor for the HiddenNeuron sets the slope of it's activation function and the number of it's inputs
HiddenNeuron(int, int, int) - Constructor for class HiddenNeuron
This constructor for the HiddenNeuron sets the slope of it's activation function, the number of it's inputs and the number of delays

I

InputNeuron - class InputNeuron.
InputNeuron class is used to create Input Neurons
InputNeuron(int) - Constructor for class InputNeuron
This constructor for the InputNeuron sets the slope of it's activation function
InputNeuron(int, int) - Constructor for class InputNeuron
This constructor for the InputNeuron sets the slope of it's activation function and the number of delays
initialise() - Method in class MLP
Initialises the network with data
initialise() - Method in class RNN
Initialises the network with data
initialise() - Method in class TDNN
Initialises the network with data
input(double) - Method in class ContextNeuron
Set the new input value and adds the old value to the memory
input(double) - Method in class InputNeuron
Sets the new input Value

M

MLP - class MLP.
The MLP class is used to create Multilayer Perceptron networks
MLP(int, int, int, double, double, int, String) - Constructor for class MLP
This constructor for the MLP specifies the number of Input Neurons and Hidden Neurons the slope of their activation functions, the learning rate, the momentum, the number of training epochs and set a network topology id for use with log files.
main(String[]) - Static method in class Test
 

O

OutputFile - class OutputFile.
The OutputFile class creates log files
OutputFile(String) - Constructor for class OutputFile
Constructor for the OutputFile
OutputNeuron - class OutputNeuron.
The OutputNeuron class creates Output Neurons
OutputNeuron(int, int) - Constructor for class OutputNeuron
This constructor for the OutputNeuron sets the slope of it's activation function and the number of it's inputs

R

RNN - class RNN.
The RNN class is based on the Elman Network, which has recurrency at the Hidden Layer.
RNN(int, int, double, int, double, double, int, String) - Constructor for class RNN
This constructor for the RNN specifies the number of Input Neurons and Hidden Neurons, the memory depth, the slope of their activation functions, the learning rate, the momentum, the number of training, epochs and set a network topology id for use with log files.

S

Synapse - class Synapse.
The Synapse class is used to connect Neurons together.
Synapse(Neuron, Neuron) - Constructor for class Synapse
This constructor for the Synapse connects two Neurons together, and sets it's weight to a random value
scale() - Method in class DataProcessor
Scales it's inputs array between 0 and 1
sendToHidden() - Method in class RNN
Transfers the weighted values to the Hidden Neurons
sendToOutput() - Method in class RNN
Transfers the weighted values to the Output Neuron
setWeight(int) - Method in class Synapse
Sets the Synapse weight

T

TDNN - class TDNN.
The TDNN class is used to create Time Delay Neural Networks
TDNN(int, int, int, int, double, double, int, String) - Constructor for class TDNN
This constructor for the TDNN specifies the number of Input Neurons and Hidden Neurons, the number of delays, the slope of their activation functions, the learning rate, the momentum, the number of training, epochs and set a network topology id for use with log files.
Test - class Test.
The Test class is the main class.
Test(String, String, String, String, String, String) - Constructor for class Test
This constructor for the Test creates training, testing and validating data sets
testMLP() - Method in class Test
Trains, tests and validates the MLP networks
testRNN() - Method in class Test
Trains, tests and validates the RNN networks
testTDNN() - Method in class Test
Trains, tests and validates the TDNN networks
transferValue() - Method in class ContextSynapse
Transfers a weighted value from the input Neuron to the output Neuron
transferValue() - Method in class Synapse
Transfers a weighted value from the input Neuron to the output Neuron

W

writeToFile(String) - Method in class OutputFile
Writes data to the file

X

XMLParser - class XMLParser.
The XMLParser class is used to parse XML documents and create W3C DOM objects
XMLParser(String) - Constructor for class XMLParser
This constructor for the XMLParserbuilds a new Document from an XML file

A B C D G H I M O R S T W X

Copyright © 2004 Raymond McBride. All Rights Reserved.