Package | Description |
---|---|
org.neuroph.core |
Provides base classes and basic building components for neural networks.
|
org.neuroph.core.events |
Provides neural network learning events system
|
org.neuroph.nnet.comp |
Provides components for the specific neural network models.
|
org.neuroph.nnet.comp.layer |
Provides various specific layer types
|
org.neuroph.nnet.comp.neuron |
Provides various specific neuron types
|
org.neuroph.nnet.learning |
Provides implementations of specific neural network learning algorithms.
|
org.neuroph.util |
Provides various utility classes for creating neural networks,
type codes, parsing vectors, etc.
|
org.neuroph.util.random |
Provides weights randomization techniques
|
Modifier and Type | Field and Description |
---|---|
protected Neuron |
Connection.fromNeuron
From neuron for this connection (source neuron).
|
protected Neuron |
Connection.toNeuron
To neuron for this connection (target, destination neuron)
This connection is input connection for to neuron.
|
Modifier and Type | Field and Description |
---|---|
protected List<Neuron> |
Layer.neurons
Collection of neurons in this layer
|
Modifier and Type | Method and Description |
---|---|
Neuron |
Connection.getFromNeuron()
Gets from neuron for this connection
|
Neuron |
Layer.getNeuronAt(int index)
Returns neuron at specified index position in this layer
|
Neuron |
Connection.getToNeuron()
Gets to neuron for this connection
|
Modifier and Type | Method and Description |
---|---|
List<Neuron> |
NeuralNetwork.getInputNeurons()
Returns input neurons
|
List<Neuron> |
Layer.getNeurons()
Returns array neurons in this layer as array
|
List<Neuron> |
NeuralNetwork.getOutputNeurons()
Returns output neurons
|
Iterator<Neuron> |
Layer.iterator() |
Modifier and Type | Method and Description |
---|---|
void |
Neuron.addInputConnection(Neuron fromNeuron)
Adds input connection from specified neuron.
|
void |
Neuron.addInputConnection(Neuron fromNeuron,
double weightVal)
Adds input connection with the given weight, from given neuron
|
void |
Layer.addNeuron(int index,
Neuron neuron)
Adds specified neuron to this layer,at specified index position
Throws IllegalArgumentException if neuron is null, or index is
illegal value (index<0 or index>neuronsCount)
|
void |
Layer.addNeuron(Neuron neuron)
Adds specified neuron to this layer
|
void |
NeuralNetwork.createConnection(Neuron fromNeuron,
Neuron toNeuron,
double weightVal)
Creates connection with specified weight value between specified neurons
|
Connection |
Neuron.getConnectionFrom(Neuron fromNeuron)
Gets input connection from the specified neuron * @param fromNeuron
neuron connected to this neuron as input
|
boolean |
Neuron.hasInputConnectionFrom(Neuron neuron)
Check the connection from neuron
|
boolean |
Neuron.hasOutputConnectionTo(Neuron toNeuron)
Check the connection to neuron
|
int |
Layer.indexOf(Neuron neuron)
Returns the index position in layer for the specified neuron
|
void |
Neuron.removeInputConnectionFrom(Neuron fromNeuron)
Removes input connection which is connected to specified neuron
|
void |
Layer.removeNeuron(Neuron neuron)
Removes neuron from layer
|
void |
Neuron.removeOutputConnectionTo(Neuron toNeuron) |
void |
Layer.setNeuron(int index,
Neuron neuron)
Sets (replace) the neuron at specified position in layer
|
Modifier and Type | Method and Description |
---|---|
void |
NeuralNetwork.setInputNeurons(List<Neuron> inputNeurons)
Sets input neurons
|
void |
NeuralNetwork.setOutputNeurons(List<Neuron> outputNeurons)
Sets output neurons
|
Constructor and Description |
---|
Connection(Neuron fromNeuron,
Neuron toNeuron)
Creates a new connection between specified neurons with random weight
|
Connection(Neuron fromNeuron,
Neuron toNeuron,
double weightVal)
Creates a new connection to specified neuron with specified weight value
|
Connection(Neuron fromNeuron,
Neuron toNeuron,
Weight weight)
Creates a new connection to specified neuron with specified weight object
|
Constructor and Description |
---|
NeuralNetworkEvent(Neuron source,
NeuralNetworkEvent.Type eventType) |
Constructor and Description |
---|
DelayedConnection(Neuron fromNeuron,
Neuron toNeuron,
double weightVal,
int delay)
Creates an instance of delayed connection to cpecified neuron and
with specified weight
|
Modifier and Type | Method and Description |
---|---|
Neuron |
FeatureMapLayer.getNeuronAt(int x,
int y)
Returns neuron at specified position in this layer
|
Neuron |
FeatureMapsLayer.getNeuronAt(int x,
int y,
int mapIndex)
Returns neuron instance at specified (x, y) position at specified feature map layer
|
Modifier and Type | Class and Description |
---|---|
class |
BiasNeuron
Neuron with constant high output (1), used as bias
|
class |
CompetitiveNeuron
Provides neuron behaviour specific for competitive neurons which are used in
competitive layers, and networks with competitive learning.
|
class |
DelayedNeuron
Provides behaviour for neurons with delayed output.
|
class |
InputNeuron
Provides input neuron behaviour - neuron with input extranaly set, which just
transfer that input to output without change.
|
class |
InputOutputNeuron
Provides behaviour specific for neurons which act as input and the output
neurons within the same layer.
|
class |
ThresholdNeuron
Provides behaviour for neurons with threshold.
|
Modifier and Type | Method and Description |
---|---|
void |
BiasNeuron.addInputConnection(Neuron fromNeuron) |
void |
BiasNeuron.addInputConnection(Neuron fromNeuron,
double weightVal) |
Modifier and Type | Method and Description |
---|---|
protected double |
ConvolutionalBackpropagation.calculateHiddenNeuronError(Neuron neuron) |
protected double |
BackPropagation.calculateHiddenNeuronError(Neuron neuron)
Calculates and returns the neuron's error (neuron's delta) for the given neuron param
|
void |
ResilientPropagation.calculateWeightChanges(Neuron neuron)
Calculate and sum gradients for each neuron's weight, the actual weight update is done in batch mode.
|
void |
QuickPropagation.calculateWeightChanges(Neuron neuron) |
void |
PerceptronLearning.calculateWeightChanges(Neuron neuron)
This method implements weights update procedure for the single neuron
In addition to weights change in LMS it applies change to neuron's threshold
|
void |
MomentumBackpropagation.calculateWeightChanges(Neuron neuron)
This method implements weights update procedure for the single neuron for
the back propagation with momentum factor
|
void |
ManhattanPropagation.calculateWeightChanges(Neuron neuron) |
protected void |
LMS.calculateWeightChanges(Neuron neuron)
This method calculates weights changes for the single neuron.
|
protected void |
UnsupervisedHebbianLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
OutstarLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
OjaLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
InstarLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
GeneralizedHebbianLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
BinaryHebbianLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
AntiHebbianLearning.updateNeuronWeights(Neuron neuron)
This method implements weights update procedure for the single neuron
|
protected void |
SupervisedHebbianLearning.updateNeuronWeights(Neuron neuron,
double desiredOutput)
This method implements weights update procedure for the single neuron
|
Modifier and Type | Method and Description |
---|---|
static Neuron |
NeuronFactory.createNeuron(NeuronProperties neuronProperties)
Creates and returns neuron instance according to the given specification in neuronProperties.
|
Modifier and Type | Method and Description |
---|---|
static void |
ConnectionFactory.createConnection(Neuron fromNeuron,
Layer toLayer)
Creates connectivity between specified neuron and all neurons in specified layer
|
static void |
ConnectionFactory.createConnection(Neuron fromNeuron,
Neuron toNeuron)
Creates connection between two specified neurons
|
static void |
ConnectionFactory.createConnection(Neuron fromNeuron,
Neuron toNeuron,
double weightVal)
Creates connection between two specified neurons
|
static void |
ConnectionFactory.createConnection(Neuron fromNeuron,
Neuron toNeuron,
double weightVal,
int delay) |
static void |
ConnectionFactory.createConnection(Neuron fromNeuron,
Neuron toNeuron,
Weight weight)
Creates connection between two specified neurons
|
Constructor and Description |
---|
NeuronProperties(Class<? extends Neuron> neuronClass) |
NeuronProperties(Class<? extends Neuron> neuronClass,
Class<? extends InputFunction> inputFunctionClass,
Class<? extends TransferFunction> transferFunctionClass) |
NeuronProperties(Class<? extends Neuron> neuronClass,
Class<? extends TransferFunction> transferFunctionClass) |
NeuronProperties(Class<? extends Neuron> neuronClass,
TransferFunctionType transferFunctionType) |
Modifier and Type | Method and Description |
---|---|
protected void |
WeightsRandomizer.randomize(Neuron neuron)
Iterates and randomizes all connection weights in specified neuron
|
protected void |
HeZhangRenSunUniformWeightsRandomizer.randomize(Neuron neuron)
"He" uniform distribution [-limit, limit] where limit is 3 * sqrt(2 / fan in)
|
protected void |
DistortRandomizer.randomize(Neuron neuron)
Iterate all layers, neurons and connection weight and apply distort randomization
|
Copyright © 2019 Neuroph Project. All rights reserved.