Neural Network Software
A B C E G H I L M N R S W

A

ActivationFunction - interface com.practicalstudies.neural.ActivationFunction.
Encapsulates the calculation of Node activation or output based on the weighted input to the Node.
Adder - interface com.practicalstudies.neural.Adder.
Interface representing a method weighting the inputs to a Node as represented by a series of Node-WeightedConnection pairs.
addInputConnection(WeightedConnection) - Method in class com.practicalstudies.neural.InputNode
As an InputNode by definition has no input connections, an attempt to call this method will result in an UnsupportedOperationException.
addInputConnection(WeightedConnection) - Method in class com.practicalstudies.neural.Neuron
Add an input connection to this neuron.
addInputConnection(WeightedConnection) - Method in interface com.practicalstudies.neural.Node
Add an input connection to this node.
addOutputConnection(WeightedConnection) - Method in class com.practicalstudies.neural.InputNode
 
addOutputConnection(WeightedConnection) - Method in class com.practicalstudies.neural.Neuron
Add an output connection to this neuron.
addOutputConnection(WeightedConnection) - Method in interface com.practicalstudies.neural.Node
Add an output connection to this node.
adjustConnectionWeight(WeightedConnection) - Method in class com.practicalstudies.neural.BackpropagationWeightAdjustment
Adjusts the weight of a WeightedConnection based on the backpropagation algorithm.
adjustConnectionWeight(WeightedConnection) - Method in interface com.practicalstudies.neural.WeightAdjustmentFunction
Any implementation of this method should call both setWeight and setLastWeightChange on the WeightedConnection object.
adjustWeight() - Method in class com.practicalstudies.neural.WeightedConnection
Called to adjust the connection weight of this connection if some learning algorithm is in place.

B

BackpropagationNode - class com.practicalstudies.neural.BackpropagationNode.
Represents a neural network node suitable for backpropagation training.
BackpropagationNode(Adder, ActivationFunction, ErrorFunction) - Constructor for class com.practicalstudies.neural.BackpropagationNode
 
BackpropagationTest - class com.practicalstudies.neural.test.BackpropagationTest.
Simple test class for the class library.
BackpropagationTest() - Constructor for class com.practicalstudies.neural.test.BackpropagationTest
 
BackpropagationWeightAdjustment - class com.practicalstudies.neural.BackpropagationWeightAdjustment.
Encapsulates the function for adjusting connection weights for backpropagation training.
BackpropagationWeightAdjustment() - Constructor for class com.practicalstudies.neural.BackpropagationWeightAdjustment
Constructor using a default learning rate of 0.5 and a default momentum coefficient of 0.5.
BackpropagationWeightAdjustment(double, double) - Constructor for class com.practicalstudies.neural.BackpropagationWeightAdjustment
Constructor using specified learning rate and momentum coefficient values.

C

calculateActivation() - Method in class com.practicalstudies.neural.Neuron
Calculate the output of this neuron.
calculateActivation() - Method in class com.practicalstudies.neural.BackpropagationNode
Calculate the output of this BackpropagationNode.
calculateActivation(double) - Method in class com.practicalstudies.neural.HyperbolicTangentFunction
Calculates neuron activation as the hyperbolic tangent function a*tanh(bx), equivalent to 2/(1 + e-(bv)) - 1 which returns values in [-1,1], with a value of 0 at the value specified by the threshold.
calculateActivation(double) - Method in class com.practicalstudies.neural.LogisticFunction
Calculates neuron activation as the logistic function 1/(1 + e-(steepness*input-threshold)), which returns values in [0,1], with a value of 0.5 at the value specified by the threshold.
calculateActivation(double) - Method in interface com.practicalstudies.neural.ActivationFunction
Calculates Node activation.
calculateError(double, double) - Method in class com.practicalstudies.neural.HyperbolicTangentFunction
Calculate the error in Neuron output relative to an actual value (useful for neurons in an output layer).
calculateError(double, double) - Method in class com.practicalstudies.neural.LogisticFunction
Calculate the error in Neuron output relative to an actual value (useful for neurons in an output layer).
calculateError(double, double) - Method in interface com.practicalstudies.neural.ErrorFunction
Calculate the error in Neuron output relative to an actual value (useful for neurons in an output layer).
calculateError(double, List) - Method in class com.practicalstudies.neural.HyperbolicTangentFunction
Calculate the error in Neuron output based on errors observed in the outputs of Neurons which rely on the output of this Neuron.
calculateError(double, List) - Method in class com.practicalstudies.neural.LogisticFunction
Calculate the error in Neuron output based on errors observed in the outputs of Neurons which rely on the output of this Neuron.
calculateError(double, List) - Method in interface com.practicalstudies.neural.ErrorFunction
Calculate the error in Neuron output based on errors observed in the outputs of Neurons which rely on the output of this Neuron.
clearInputConnections() - Method in class com.practicalstudies.neural.InputNode
As an InputNode by definition has no input connections, an attempt to call this method will result in an UnsupportedOperationException.
clearInputConnections() - Method in class com.practicalstudies.neural.Neuron
Remove all input connections from this neuron.
clearInputConnections() - Method in interface com.practicalstudies.neural.Node
Remove all input connections from this node.
clearOutputConnections() - Method in class com.practicalstudies.neural.InputNode
 
clearOutputConnections() - Method in class com.practicalstudies.neural.Neuron
Remove all output connections from this neuron.
clearOutputConnections() - Method in interface com.practicalstudies.neural.Node
Remove all output connections from this node.
com.practicalstudies.neural - package com.practicalstudies.neural
Neural network class library.
com.practicalstudies.neural.test - package com.practicalstudies.neural.test
Test classes for the neural network class library.

E

ErrorFunction - interface com.practicalstudies.neural.ErrorFunction.
Represents the calculation of the error in Neuron output, either relative to an actual value, or based on error in the outputs of Neurons to which the Neuron feeds.

G

getError(double) - Method in class com.practicalstudies.neural.BackpropagationNode
Returns the error in the output of this BackpropagationNode.
getInputConnection(int) - Method in class com.practicalstudies.neural.InputNode
As an InputNode by definition has no input connections, an attempt to call this method will result in an UnsupportedOperationException.
getInputConnection(int) - Method in class com.practicalstudies.neural.Neuron
Return a specific input connection to this neuron.
getInputConnection(int) - Method in interface com.practicalstudies.neural.Node
Return a specific input connection to this node.
getInputConnections() - Method in class com.practicalstudies.neural.InputNode
As an InputNode by definition has no input connections, returns an empty List.
getInputConnections() - Method in class com.practicalstudies.neural.Neuron
Returns a List of WeightedConnections which are the input connections to this neuron.
getInputConnections() - Method in interface com.practicalstudies.neural.Node
Returns a List of WeightedConnections which are the input connections to this node.
getInputNode() - Method in class com.practicalstudies.neural.WeightedConnection
 
getLastWeightChange() - Method in class com.practicalstudies.neural.WeightedConnection
Returns the last change in weight of this function during network learning.
getLearningRate() - Method in class com.practicalstudies.neural.BackpropagationWeightAdjustment
 
getMomentumCoefficient() - Method in class com.practicalstudies.neural.BackpropagationWeightAdjustment
 
getNumberInputConnections() - Method in class com.practicalstudies.neural.InputNode
As an InputNode by definition has no input connections, returns 0.
getNumberInputConnections() - Method in class com.practicalstudies.neural.Neuron
Returns the number of input connections to this neuron.
getNumberInputConnections() - Method in interface com.practicalstudies.neural.Node
Returns the number of input connections to this node.
getNumberOutputConnections() - Method in class com.practicalstudies.neural.InputNode
 
getNumberOutputConnections() - Method in class com.practicalstudies.neural.Neuron
Returns the number of output connections from this neuron.
getNumberOutputConnections() - Method in interface com.practicalstudies.neural.Node
Returns the number of output connections from this node.
getOutput() - Method in class com.practicalstudies.neural.InputNode
Returns an input datum.
getOutput() - Method in class com.practicalstudies.neural.Neuron
Return the output of this neuron.
getOutput() - Method in interface com.practicalstudies.neural.Node
Return the output of this node.
getOutputConnection(int) - Method in class com.practicalstudies.neural.InputNode
 
getOutputConnection(int) - Method in class com.practicalstudies.neural.Neuron
Return a specific output connection to this neuron.
getOutputConnection(int) - Method in interface com.practicalstudies.neural.Node
Return a specific output connection to this node.
getOutputConnections() - Method in class com.practicalstudies.neural.InputNode
 
getOutputConnections() - Method in class com.practicalstudies.neural.Neuron
Returns a List of WeightedConnections which are the output connections to this neuron.
getOutputConnections() - Method in interface com.practicalstudies.neural.Node
Returns a List of WeightedConnections which are the output connections to this node.
getOutputNode() - Method in class com.practicalstudies.neural.WeightedConnection
 
getSteepness() - Method in class com.practicalstudies.neural.HyperbolicTangentFunction
 
getSteepness() - Method in class com.practicalstudies.neural.LogisticFunction
 
getThreshold() - Method in class com.practicalstudies.neural.LogisticFunction
 
getWeight() - Method in class com.practicalstudies.neural.WeightedConnection
Returns the weight of this connection.
getWeightAdjustmentFunction() - Method in class com.practicalstudies.neural.WeightedConnection
 
getWeightedInput(List) - Method in class com.practicalstudies.neural.LinearAdder
Returns a linear combination of the weighted inputs of the input WeightedConnections to a Node
getWeightedInput(List) - Method in interface com.practicalstudies.neural.Adder
Function returning the weighted inputs of the input WeightedConnections to a Node

H

HyperbolicTangentFunction - class com.practicalstudies.neural.HyperbolicTangentFunction.
Encapsulates the calculation of Node activation or output based on the hyperbolic tangent function, and of the error in output for the backpropagation training method using this function.
HyperbolicTangentFunction() - Constructor for class com.practicalstudies.neural.HyperbolicTangentFunction
Constructor for a function with a steepness of 1.
HyperbolicTangentFunction(double) - Constructor for class com.practicalstudies.neural.HyperbolicTangentFunction
Constructor for a function with a specified steepness.

I

InputNode - class com.practicalstudies.neural.InputNode.
Represents an input node to a network.
InputNode() - Constructor for class com.practicalstudies.neural.InputNode
 

L

LinearAdder - class com.practicalstudies.neural.LinearAdder.
Class representing a linear combination of the outputs of a group of Nodes weighted according to weights contained in associated WeightedConnections.
LinearAdder() - Constructor for class com.practicalstudies.neural.LinearAdder
 
LogisticFunction - class com.practicalstudies.neural.LogisticFunction.
Encapsulates the calculation of Node activation or output based on the logistic function, and of the error in output for the backpropagation training method using this function.
LogisticFunction() - Constructor for class com.practicalstudies.neural.LogisticFunction
Constructor for a function with a threshold of 0 and a steepness of 1.
LogisticFunction(double) - Constructor for class com.practicalstudies.neural.LogisticFunction
Constructor for a function with specified threshold and steepness values.

M

main(String[]) - Static method in class com.practicalstudies.neural.test.BackpropagationTest
 

N

Neuron - class com.practicalstudies.neural.Neuron.
A framework class representing a neuron in an artificial neural network.
Neuron(Adder, ActivationFunction) - Constructor for class com.practicalstudies.neural.Neuron
Constructor.
Node - interface com.practicalstudies.neural.Node.
Interface representing the minimal functionality of a node in a neural network.

R

randomizeWeight() - Method in class com.practicalstudies.neural.WeightedConnection
Assign a random value to the connection weight of this connection.

S

setInputNode(Node) - Method in class com.practicalstudies.neural.WeightedConnection
 
setLastWeightChange(double) - Method in class com.practicalstudies.neural.WeightedConnection
Specify the last change in weight of this function.
setLearningRate(double) - Method in class com.practicalstudies.neural.BackpropagationWeightAdjustment
Specify the learning rate.
setMomentumCoefficient(double) - Method in class com.practicalstudies.neural.BackpropagationWeightAdjustment
Specify the momentum coefficient, which is the fraction of the previous weight change to apply during the current adjustment.
setOutput(double) - Method in class com.practicalstudies.neural.InputNode
Specify the value to be returned by getOutput.
setOutputNode(Node) - Method in class com.practicalstudies.neural.WeightedConnection
 
setSteepness(double) - Method in class com.practicalstudies.neural.HyperbolicTangentFunction
Specify the steepness parameter of the logistic function.
setSteepness(double) - Method in class com.practicalstudies.neural.LogisticFunction
Specify the steepness parameter of the logistic function.
setThreshold(double) - Method in class com.practicalstudies.neural.LogisticFunction
Specify a threshold value, which determines where the logistic function is centered.
setWeight(double) - Method in class com.practicalstudies.neural.WeightedConnection
Specify the weight of this connection.
setWeightAdjustmentFunction(WeightAdjustmentFunction) - Method in class com.practicalstudies.neural.WeightedConnection
 

W

WeightAdjustmentFunction - interface com.practicalstudies.neural.WeightAdjustmentFunction.
Encapsulates an algorithm to adjust the connection weight of a WeightedConnection, typically to support network learning.
WeightedConnection - class com.practicalstudies.neural.WeightedConnection.
Represents a connection between Nodes.
WeightedConnection() - Constructor for class com.practicalstudies.neural.WeightedConnection
Constructor.
WeightedConnection(Node, Node) - Constructor for class com.practicalstudies.neural.WeightedConnection
Constructor which allows assignment of input and output Nodes to the connection.

A B C E G H I L M N R S W
PracticalStudies.com