NeuralNetwork (Core 2.96 API)

java.lang.Object
- org.neuroph.core.NeuralNetwork<L>

All Implemented Interfaces:: Serializable

Direct Known Subclasses:: Adaline, BAM, CompetitiveNetwork, ConvolutionalNetwork, ElmanNetwork, Hopfield, Instar, JordanNetwork, Kohonen, MaxNet, MultiLayerPerceptron, NeuroFuzzyPerceptron, Outstar, Perceptron, RBFNetwork, RectifierNeuralNetwork, SupervisedHebbianNetwork, UnsupervisedHebbianNetwork

public class NeuralNetwork<L extends LearningRule>
extends Object
implements Serializable

 Base class for artificial neural networks. It provides generic structure and functionality
 for the neural networks. Neural network contains a collection of neuron layers and learning rule.
 Custom neural networks are created by deriving from this class, creating layers of interconnected network specific neurons,
 and setting network specific learning rule.

Author:: Zoran Sevarac
See Also:: Layer, LearningRule, Serialized Form

Field Summary

Fields
Modifier and Type Field and Description

protected double[] outputBuffer
Neural network output buffer

Fields
Modifier and Type	Field and Description
`protected double[]`	`outputBuffer` Neural network output buffer

Constructor Summary

Constructors
Constructor and Description

NeuralNetwork()
Creates an instance of empty neural network.

Constructors
Constructor and Description
`NeuralNetwork()` Creates an instance of empty neural network.

Method Summary

All Methods Static Methods Instance Methods Concrete Methods Deprecated Methods
Modifier and Type	Method and Description
`void`	`addLayer(int index, Layer layer)` Adds layer to specified index position in network
`void`	`addLayer(Layer layer)` Adds layer to neural network
`void`	`addListener(NeuralNetworkEventListener listener)`
`void`	`addPlugin(PluginBase plugin)` Adds plugin to neural network
`void`	`calculate()` Performs calculation on whole network
`void`	`createConnection(Neuron fromNeuron, Neuron toNeuron, double weightVal)` Creates connection with specified weight value between specified neurons
`static NeuralNetwork`	`createFromFile(File file)` Loads and return s neural network instance from specified file
`static NeuralNetwork`	`createFromFile(String filePath)`
`void`	`fireNetworkEvent(NeuralNetworkEvent evt)`
`List<Neuron>`	`getInputNeurons()` Returns input neurons
`int`	`getInputsCount()` Gets number of input neurons
`String`	`getLabel()` Get network label
`Layer`	`getLayerAt(int index)` Returns layer at specified index
`List<Layer>`	`getLayers()` Returns layers array
`int`	`getLayersCount()` Returns number of layers in network
`L`	`getLearningRule()` Returns the learning algorithm of this network
`NeuralNetworkType`	`getNetworkType()` Returns type of this network
`double[]`	`getOutput()` Returns network output vector.
`String[]`	`getOutputLabels()`
`List<Neuron>`	`getOutputNeurons()` Returns output neurons
`int`	`getOutputsCount()`
`<T extends PluginBase> T`	`getPlugin(Class<T> pluginClass)` Returns the requested plugin
`Double[]`	`getWeights()` Returns all network weights as an double array
`int`	`indexOf(Layer layer)` Returns index position of the specified layer
`boolean`	`isEmpty()`
`void`	`learn(DataSet trainingSet)` Learn the specified training set
`void`	`learn(DataSet trainingSet, L learningRule)` Learn the specified training set, using specified learning rule
`static NeuralNetwork`	`load(InputStream inputStream)` Loads neural network from the specified InputStream.
`static NeuralNetwork`	`load(String filePath)` Deprecated. Use createFromFile method instead
`void`	`pauseLearning()` Pause the learning - puts learning thread in ca state.
`void`	`randomizeWeights()` Randomizes connection weights for the whole network
`void`	`randomizeWeights(double minWeight, double maxWeight)` Randomizes connection weights for the whole network within specified value range
`void`	`randomizeWeights(Random random)` Randomizes connection weights for the whole network using specified random generator
`void`	`randomizeWeights(WeightsRandomizer randomizer)` Randomizes connection weights for the whole network using specified randomizer
`void`	`removeLayer(Layer layer)` Removes specified layer from network
`void`	`removeLayerAt(int index)` Removes layer at specified index position from net
`void`	`removeListener(NeuralNetworkEventListener listener)`
`void`	`removePlugin(Class pluginClass)` Removes the plugin with specified name
`void`	`reset()` Resets the activation levels for whole network
`void`	`resumeLearning()` Resumes paused learning - notifies the learning rule to continue
`void`	`save(String filePath)` Saves neural network into the specified file.
`void`	`setInput(double... inputVector)` Sets network input.
`void`	`setInputNeurons(List<Neuron> inputNeurons)` Sets input neurons
`void`	`setLabel(String label)` Set network label
`void`	`setLearningRule(L learningRule)` Sets learning algorithm for this network
`void`	`setNetworkType(NeuralNetworkType type)` Sets type for this network
`void`	`setOutputLabels(String[] labels)` Sets labels for output neurons
`void`	`setOutputNeurons(List<Neuron> outputNeurons)` Sets output neurons
`void`	`setWeights(double[] weights)` Sets network weights from the specified double array
`void`	`stopLearning()` Stops learning
`String`	`toString()`

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

- Field Detail
  - outputBuffer
```
protected double[] outputBuffer
```
    Neural network output buffer
- Constructor Detail
  - NeuralNetwork
```
public NeuralNetwork()
```
    Creates an instance of empty neural network.
- Method Detail
  - addLayer
```
public void addLayer(Layer layer)
```
    Adds layer to neural network
    
    Parameters:
    
    layer - layer to add
  - addLayer
```
public void addLayer(int index,
                     Layer layer)
```
    Adds layer to specified index position in network
    
    Parameters:
    
    index - index position to add layer
    
    layer - layer to add
  - removeLayer
```
public void removeLayer(Layer layer)
```
    Removes specified layer from network
    
    Parameters:
    
    layer - layer to remove
    
    Throws:
    
    Exception
  - removeLayerAt
```
public void removeLayerAt(int index)
```
    Removes layer at specified index position from net
    
    Parameters:
    
    index - int value represents index postion of layer which should be removed
  - getLayers
```
public List<Layer> getLayers()
```
    Returns layers array
    
    Returns:
    
    array of layers
  - getLayerAt
```
public Layer getLayerAt(int index)
```
    Returns layer at specified index
    
    Parameters:
    
    index - layer index position
    
    Returns:
    
    layer at specified index position
  - indexOf
```
public int indexOf(Layer layer)
```
    Returns index position of the specified layer
    
    Parameters:
    
    layer - requested Layer object
    
    Returns:
    
    layer position index
  - getLayersCount
```
public int getLayersCount()
```
    Returns number of layers in network
    
    Returns:
    
    number of layes in net
  - setInput
```
public void setInput(double... inputVector)
              throws VectorSizeMismatchException
```
    Sets network input. Input is an array of double values.
    
    Parameters:
    
    inputVector - network input as double array
    
    Throws:
    
    VectorSizeMismatchException
  - getOutput
```
public double[] getOutput()
```
    Returns network output vector. Output vector is an array collection of Double values.
    
    Returns:
    
    network output vector
  - calculate
```
public void calculate()
```
    Performs calculation on whole network
  - reset
```
public void reset()
```
    Resets the activation levels for whole network
  - learn
```
public void learn(DataSet trainingSet)
```
    Learn the specified training set
    
    Parameters:
    
    trainingSet - set of training elements to learn
  - learn
```
public void learn(DataSet trainingSet,
                  L learningRule)
```
    Learn the specified training set, using specified learning rule
    
    Parameters:
    
    trainingSet - set of training elements to learn
    
    learningRule - instance of learning rule to use for learning
  - stopLearning
```
public void stopLearning()
```
    Stops learning
  - pauseLearning
```
public void pauseLearning()
```
    Pause the learning - puts learning thread in ca state. Makes sense only wen learning is done in new thread with learnInNewThread() method
  - resumeLearning
```
public void resumeLearning()
```
    Resumes paused learning - notifies the learning rule to continue
  - randomizeWeights
```
public void randomizeWeights()
```
    Randomizes connection weights for the whole network
  - randomizeWeights
```
public void randomizeWeights(double minWeight,
                             double maxWeight)
```
    Randomizes connection weights for the whole network within specified value range
  - randomizeWeights
```
public void randomizeWeights(Random random)
```
    Randomizes connection weights for the whole network using specified random generator
  - randomizeWeights
```
public void randomizeWeights(WeightsRandomizer randomizer)
```
    Randomizes connection weights for the whole network using specified randomizer
    
    Parameters:
    
    randomizer - random weight generator to use
  - getNetworkType
```
public NeuralNetworkType getNetworkType()
```
    Returns type of this network
    
    Returns:
    
    network type
  - setNetworkType
```
public void setNetworkType(NeuralNetworkType type)
```
    Sets type for this network
    
    Parameters:
    
    type - network type
  - getInputNeurons
```
public List<Neuron> getInputNeurons()
```
    Returns input neurons
    
    Returns:
    
    input neurons
  - getInputsCount
```
public int getInputsCount()
```
    Gets number of input neurons
    
    Returns:
    
    number of input neurons
  - setInputNeurons
```
public void setInputNeurons(List<Neuron> inputNeurons)
```
    Sets input neurons
    
    Parameters:
    
    inputNeurons - array of input neurons
  - getOutputNeurons
```
public List<Neuron> getOutputNeurons()
```
    Returns output neurons
    
    Returns:
    
    list of output neurons
  - getOutputsCount
```
public int getOutputsCount()
```
  - setOutputNeurons
```
public void setOutputNeurons(List<Neuron> outputNeurons)
```
    Sets output neurons
    
    Parameters:
    
    outputNeurons - output neurons collection
  - setOutputLabels
```
public void setOutputLabels(String[] labels)
```
    Sets labels for output neurons
    
    Parameters:
    
    labels - labels for output neurons
  - getOutputLabels
```
public String[] getOutputLabels()
```
  - getLearningRule
```
public L getLearningRule()
```
    Returns the learning algorithm of this network
    
    Returns:
    
    algorithm for network training
  - setLearningRule
```
public void setLearningRule(L learningRule)
```
    Sets learning algorithm for this network
    
    Parameters:
    
    learningRule - learning algorithm for this network
  - getWeights
```
public Double[] getWeights()
```
    Returns all network weights as an double array
    
    Returns:
    
    network weights as an double array
  - setWeights
```
public void setWeights(double[] weights)
```
    Sets network weights from the specified double array
    
    Parameters:
    
    weights - array of weights to set
  - isEmpty
```
public boolean isEmpty()
```
  - createConnection
```
public void createConnection(Neuron fromNeuron,
                             Neuron toNeuron,
                             double weightVal)
```
    Creates connection with specified weight value between specified neurons
    
    Parameters:
    
    fromNeuron - neuron to connect
    
    toNeuron - neuron to connect to
    
    weightVal - connection weight value
  - toString
```
public String toString()
```
    Overrides:
    
    toString in class Object
  - save
```
public void save(String filePath)
```
    Saves neural network into the specified file.
    
    Parameters:
    
    filePath - file path to save network into
  - load
```
public static NeuralNetwork load(String filePath)
```
    Deprecated. Use createFromFile method instead
    
    Loads neural network from the specified file.
    
    Parameters:
    
    filePath - file path to load network from
    
    Returns:
    
    loaded neural network as NeuralNetwork object
  - load
```
public static NeuralNetwork load(InputStream inputStream)
```
    Loads neural network from the specified InputStream.
    
    Parameters:
    
    inputStream - input stream to load network from
    
    Returns:
    
    loaded neural network as NeuralNetwork object
  - createFromFile
```
public static NeuralNetwork createFromFile(File file)
```
    Loads and return s neural network instance from specified file
    
    Parameters:
    
    file - neural network file
    
    Returns:
    
    neural network instance
  - createFromFile
```
public static NeuralNetwork createFromFile(String filePath)
```
  - addPlugin
```
public void addPlugin(PluginBase plugin)
```
    Adds plugin to neural network
    
    Parameters:
    
    plugin - neural network plugin to add
  - getPlugin
```
public <T extends PluginBase> T getPlugin(Class<T> pluginClass)
```
    Returns the requested plugin
    
    Parameters:
    
    pluginClass - class of the plugin to get
    
    Returns:
    
    instance of specified plugin class
  - removePlugin
```
public void removePlugin(Class pluginClass)
```
    Removes the plugin with specified name
    
    Parameters:
    
    pluginClass - class of the plugin to remove
  - getLabel
```
public String getLabel()
```
    Get network label
    
    Returns:
    
    network label
  - setLabel
```
public void setLabel(String label)
```
    Set network label
    
    Parameters:
    
    label - network label to set
  - addListener
```
public void addListener(NeuralNetworkEventListener listener)
```
  - removeListener
```
public void removeListener(NeuralNetworkEventListener listener)
```
  - fireNetworkEvent
```
public void fireNetworkEvent(NeuralNetworkEvent evt)
```

Class NeuralNetwork<L extends LearningRule>

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

outputBuffer

Constructor Detail

NeuralNetwork

Method Detail

addLayer

addLayer

removeLayer

removeLayerAt

getLayers

getLayerAt

indexOf

getLayersCount

setInput

getOutput

calculate

reset

learn

learn

stopLearning

pauseLearning

resumeLearning

randomizeWeights

randomizeWeights

randomizeWeights

randomizeWeights

getNetworkType

setNetworkType

getInputNeurons

getInputsCount

setInputNeurons

getOutputNeurons

getOutputsCount

setOutputNeurons

setOutputLabels

getOutputLabels

getLearningRule

setLearningRule

getWeights

setWeights

isEmpty

createConnection

toString

save

load

load

createFromFile

createFromFile

addPlugin

getPlugin

removePlugin

getLabel

setLabel

addListener

removeListener

fireNetworkEvent