org.encog.neural.networks

Class BasicNetwork

java.lang.Object

org.encog.ml.BasicML

org.encog.neural.networks.BasicNetwork

All Implemented Interfaces:

Serializable, MLClassification, MLContext, MLEncodable, MLError, MLFactory, MLInput, MLInputOutput, MLMethod, MLOutput, MLProperties, MLRegression, MLResettable, ContainsFlat

public class BasicNetwork
extends BasicML
implements ContainsFlat, MLContext, MLRegression, MLEncodable, MLResettable, MLClassification, MLError, MLFactory

This class implements a neural network. This class works in conjunction the Layer classes. Layers are added to the BasicNetwork to specify the structure of the neural network. The first layer added is the input layer, the final layer added is the output layer. Any layers added between these two layers are the hidden layers. The network structure is stored in the structure member. It is important to call: network.getStructure().finalizeStructure(); Once the neural network has been completely constructed.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_CONNECTION_LIMIT The default connection limit.
static String	TAG_BEGIN_TRAINING The property for begin training.
static String	TAG_BIAS_ACTIVATION The property for bias activation.
static String	TAG_CONNECTION_LIMIT The property for connection limit.
static String	TAG_CONTEXT_TARGET_OFFSET The property for context target offset.
static String	TAG_CONTEXT_TARGET_SIZE The property for context target size.
static String	TAG_END_TRAINING The property for end training.
static String	TAG_HAS_CONTEXT The property for has context.
static String	TAG_LAYER_CONTEXT_COUNT The property for layer context count.
static String	TAG_LAYER_COUNTS The property for layer counts.
static String	TAG_LAYER_FEED_COUNTS The property for layer feed counts.
static String	TAG_LAYER_INDEX The property for layer index.
static String	TAG_LIMIT Tag used for the connection limit.
static String	TAG_WEIGHT_INDEX The property for weight index.

Constructor Summary

Constructors

Constructor and Description
BasicNetwork() Construct an empty neural network.

Method Summary

Methods

Modifier and Type	Method and Description
void	addLayer(Layer layer) Add a layer to the neural network.
void	addWeight(int fromLayer, int fromNeuron, int toNeuron, double value) Add to a weight.
double	calculateError(MLDataSet data) Calculate the error for this neural network.
int	calculateNeuronCount() Calculate the total number of neurons in the network across all layers.
int	classify(MLData input) Classify the input into a group.
void	clearContext() Clear any data from any context layers.
Object	clone() Return a clone of this neural network.
void	compute(double[] input, double[] output) Compute the output for this network.
MLData	compute(MLData input) Compute the output for a given input to the neural network.
void	decodeFromArray(double[] encoded) Decode an array to this object.
String	dumpWeights()
void	enableConnection(int fromLayer, int fromNeuron, int toNeuron, boolean enable) Enable, or disable, a connection.
int	encodedArrayLength()
void	encodeToArray(double[] encoded) Encode the object to the specified array.
boolean	equals(BasicNetwork other, int precision) Determine if this neural network is equal to another.
boolean	equals(Object other) Compare the two neural networks.
ActivationFunction	getActivation(int layer) Get the activation function for the specified layer.
String	getFactoryArchitecture()
String	getFactoryType()
FlatNetwork	getFlat()
int	getInputCount()
double	getLayerBiasActivation(int l) Get the bias activation for the specified layer.
int	getLayerCount()
int	getLayerNeuronCount(int l) Get the neuron count.
double	getLayerOutput(int layer, int neuronNumber) Get the layer output for the specified neuron.
int	getLayerTotalNeuronCount(int l) Get the total (including bias and context) neuron cont for a layer.
int	getOutputCount()
NeuralStructure	getStructure()
double	getWeight(int fromLayer, int fromNeuron, int toNeuron) Get the weight between the two layers.
int	hashCode() Generate a hash code.
boolean	isConnected(int layer, int fromNeuron, int toNeuron) Determine if the specified connection is enabled.
boolean	isLayerBiased(int l) Determine if the specified layer is biased.
void	reset() Reset the weight matrix and the bias values.
void	reset(int seed) Reset the weight matrix and the bias values.
void	setBiasActivation(double activation) Sets the bias activation for every layer that supports bias.
void	setLayerBiasActivation(int l, double value) Set the bias activation for the specified layer.
void	setWeight(int fromLayer, int fromNeuron, int toNeuron, double value) Set the weight between the two specified neurons.
String	toString()
void	updateProperties() Update any objeccts when a property changes.
void	validateNeuron(int targetLayer, int neuron) Validate the the specified targetLayer and neuron are valid.
int	winner(MLData input) Determine the winner for the specified input.

Methods inherited from class org.encog.ml.BasicML

getProperties, getPropertyDouble, getPropertyLong, getPropertyString, setProperty, setProperty, setProperty

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Field Detail

TAG_LIMIT

public static final String TAG_LIMIT

Tag used for the connection limit.

See Also:

Constant Field Values

DEFAULT_CONNECTION_LIMIT

public static final double DEFAULT_CONNECTION_LIMIT

The default connection limit.

See Also:

Constant Field Values

TAG_CONNECTION_LIMIT

public static final String TAG_CONNECTION_LIMIT

The property for connection limit.

See Also:

Constant Field Values

TAG_BEGIN_TRAINING

public static final String TAG_BEGIN_TRAINING

The property for begin training.

See Also:

Constant Field Values

TAG_CONTEXT_TARGET_OFFSET

public static final String TAG_CONTEXT_TARGET_OFFSET

The property for context target offset.

See Also:

Constant Field Values

TAG_CONTEXT_TARGET_SIZE

public static final String TAG_CONTEXT_TARGET_SIZE

The property for context target size.

See Also:

Constant Field Values

TAG_END_TRAINING

public static final String TAG_END_TRAINING

The property for end training.

See Also:

Constant Field Values

TAG_HAS_CONTEXT

public static final String TAG_HAS_CONTEXT

The property for has context.

See Also:

Constant Field Values

TAG_LAYER_COUNTS

public static final String TAG_LAYER_COUNTS

The property for layer counts.

See Also:

Constant Field Values

TAG_LAYER_FEED_COUNTS

public static final String TAG_LAYER_FEED_COUNTS

The property for layer feed counts.

See Also:

Constant Field Values

TAG_LAYER_INDEX

public static final String TAG_LAYER_INDEX

The property for layer index.

See Also:

Constant Field Values

TAG_WEIGHT_INDEX

public static final String TAG_WEIGHT_INDEX

The property for weight index.

See Also:

Constant Field Values

TAG_BIAS_ACTIVATION

public static final String TAG_BIAS_ACTIVATION

The property for bias activation.

See Also:

Constant Field Values

TAG_LAYER_CONTEXT_COUNT

public static final String TAG_LAYER_CONTEXT_COUNT

The property for layer context count.

See Also:

Constant Field Values

Constructor Detail

BasicNetwork

public BasicNetwork()

Construct an empty neural network.

Method Detail

addLayer

public void addLayer(Layer layer)

Add a layer to the neural network. If there are no layers added this layer will become the input layer. This function automatically updates both the input and output layer references.

Parameters:

layer - The layer to be added to the network.

addWeight

public void addWeight(int fromLayer,
             int fromNeuron,
             int toNeuron,
             double value)

Add to a weight.

Parameters:

fromLayer - The from layer.

fromNeuron - The from neuron.

toNeuron - The to neuron.

value - The value to add.

calculateError

public double calculateError(MLDataSet data)

Calculate the error for this neural network. We always calculate the error using the "regression" calculator. Neural networks don't directly support classification, rather they use one-of-encoding or similar. So just using the regression calculator gives a good approximation.

Specified by:

calculateError in interface MLError

Parameters:

data - The training set.

Returns:

The error percentage.

calculateNeuronCount

public int calculateNeuronCount()

Calculate the total number of neurons in the network across all layers.

Returns:

The neuron count.

classify

public int classify(MLData input)

Classify the input into a group.

Specified by:

classify in interface MLClassification

Parameters:

input - The input data to classify.

Returns:

The group that the data was classified into.

clearContext

public void clearContext()

Clear any data from any context layers.

Specified by:

clearContext in interface MLContext

clone

public Object clone()

Return a clone of this neural network. Including structure, weights and bias values. This is a deep copy.

Overrides:

clone in class Object

Returns:

A cloned copy of the neural network.

compute

public void compute(double[] input,
           double[] output)

Compute the output for this network.

Parameters:

input - The input.

output - The output.

compute

public MLData compute(MLData input)

Compute the output for a given input to the neural network.

Specified by:

compute in interface MLRegression

Parameters:

input - The input to the neural network.

Returns:

The output from the neural network.

decodeFromArray

public void decodeFromArray(double[] encoded)

Decode an array to this object.

Specified by:

decodeFromArray in interface MLEncodable

Parameters:

encoded - The encoded array.

dumpWeights

public String dumpWeights()

Returns:

The weights as a comma separated list.

enableConnection

public void enableConnection(int fromLayer,
                    int fromNeuron,
                    int toNeuron,
                    boolean enable)

Enable, or disable, a connection.

Parameters:

fromLayer - The layer that contains the from neuron.

fromNeuron - The source neuron.

toNeuron - The target connection.

enable - True to enable, false to disable.

encodedArrayLength

public int encodedArrayLength()

Specified by:

encodedArrayLength in interface MLEncodable

Returns:

The length of an encoded array.

encodeToArray

public void encodeToArray(double[] encoded)

Encode the object to the specified array.

Specified by:

encodeToArray in interface MLEncodable

Parameters:

encoded - The array.

equals

public boolean equals(Object other)

Compare the two neural networks. For them to be equal they must be of the same structure, and have the same matrix values.

Overrides:

equals in class Object

Parameters:

other - The other neural network.

Returns:

True if the two networks are equal.

equals

public boolean equals(BasicNetwork other,
             int precision)

Determine if this neural network is equal to another. Equal neural networks have the same weight matrix and bias values, within a specified precision.

Parameters:

other - The other neural network.

precision - The number of decimal places to compare to.

Returns:

True if the two neural networks are equal.

getActivation

public ActivationFunction getActivation(int layer)

Get the activation function for the specified layer.

Parameters:

layer - The layer.

Returns:

The activation function.

getFlat

public FlatNetwork getFlat()

Specified by:

getFlat in interface ContainsFlat

Returns:

The flat network associated with this neural network.

getInputCount

public int getInputCount()

Specified by:

getInputCount in interface MLInput

Returns:

The input.

getLayerBiasActivation

public double getLayerBiasActivation(int l)

Get the bias activation for the specified layer.

Parameters:

l - The layer.

Returns:

The bias activation.

getLayerCount

public int getLayerCount()

Returns:

The layer count.

getLayerNeuronCount

public int getLayerNeuronCount(int l)

Get the neuron count.

Parameters:

l - The layer.

Returns:

The neuron count.

getLayerOutput

public double getLayerOutput(int layer,
                    int neuronNumber)

Get the layer output for the specified neuron.

Parameters:

layer - The layer.

neuronNumber - The neuron number.

Returns:

The output from the last call to compute.

getLayerTotalNeuronCount

public int getLayerTotalNeuronCount(int l)

Get the total (including bias and context) neuron cont for a layer.

Parameters:

l - The layer.

Returns:

The count.

getOutputCount

public int getOutputCount()

Specified by:

getOutputCount in interface MLOutput

Returns:

The output count.

getStructure

public NeuralStructure getStructure()

Returns:

Get the structure of the neural network. The structure allows you to quickly obtain synapses and layers without traversing the network.

getWeight

public double getWeight(int fromLayer,
               int fromNeuron,
               int toNeuron)

Get the weight between the two layers.

Parameters:

fromLayer - The from layer.

fromNeuron - The from neuron.

toNeuron - The to neuron.

Returns:

The weight value.

hashCode

public int hashCode()

Generate a hash code.

Overrides:

hashCode in class Object

Returns:

THe hash code.

isConnected

public boolean isConnected(int layer,
                  int fromNeuron,
                  int toNeuron)

Determine if the specified connection is enabled.

Parameters:

layer - The layer to check.

fromNeuron - The source neuron.

toNeuron - THe target neuron.

Returns:

True, if the connection is enabled, false otherwise.

isLayerBiased

public boolean isLayerBiased(int l)

Determine if the specified layer is biased.

Parameters:

l - The layer number.

Returns:

True, if the layer is biased.

reset

public void reset()

Reset the weight matrix and the bias values. This will use a Nguyen-Widrow randomizer with a range between -1 and 1. If the network does not have an input, output or hidden layers, then Nguyen-Widrow cannot be used and a simple range randomize between -1 and 1 will be used.

Specified by:

reset in interface MLResettable

reset

public void reset(int seed)

Reset the weight matrix and the bias values. This will use a RangeRandomizer with a range between -1 and 1.

Specified by:

reset in interface MLResettable

Parameters:

seed - The seed value.

setBiasActivation

public void setBiasActivation(double activation)

Sets the bias activation for every layer that supports bias. Make sure that the network structure has been finalized before calling this method.

Parameters:

activation - THe new activation.

setLayerBiasActivation

public void setLayerBiasActivation(int l,
                          double value)

Set the bias activation for the specified layer.

Parameters:

l - The layer to use.

value - The bias activation.

setWeight

public void setWeight(int fromLayer,
             int fromNeuron,
             int toNeuron,
             double value)

Set the weight between the two specified neurons. The bias neuron is always the last neuron on a layer.

Parameters:

fromLayer - The from layer.

fromNeuron - The from neuron.

toNeuron - The to neuron.

value - The to value.

toString

public String toString()

Overrides:

toString in class Object

updateProperties

public void updateProperties()

Update any objeccts when a property changes.

Specified by:

updateProperties in interface MLProperties

Specified by:

updateProperties in class BasicML

validateNeuron

public void validateNeuron(int targetLayer,
                  int neuron)

Validate the the specified targetLayer and neuron are valid.

Parameters:

targetLayer - The target layer.

neuron - The target neuron.

winner

public int winner(MLData input)

Determine the winner for the specified input. This is the number of the winning neuron.

Parameters:

input - The input patter to present to the neural network.

Returns:

The winning neuron.

getFactoryType

public String getFactoryType()

Specified by:

getFactoryType in interface MLFactory

Returns:

The Encog factory type code.

getFactoryArchitecture

public String getFactoryArchitecture()

Specified by:

getFactoryArchitecture in interface MLFactory

Returns:

The Encog architecture code.