C# Класс Encog.Neural.Networks.BasicNetwork

Наследование: Encog.ML.BasicML, IMLMethod, IContainsFlat, IMLContext, IMLRegression, IMLInputOutput, IMLInput, IMLOutput, IMLEncodable, IMLResettable, IMLClassification, IMLError

Показать файл Открыть проект Примеры использования класса

Открытые методы

Метод	Описание
AddLayer ( ILayer layer ) : void	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.
AddWeight ( int fromLayer, int fromNeuron, int toNeuron, double v ) : void	Add to a weight.
BasicNetwork ( ) : System	Construct an empty neural network.
CalculateError ( IMLDataSet data ) : double	Calculate the error for this neural network.
CalculateNeuronCount ( ) : int	Calculate the total number of neurons in the network across all layers.
Classify ( IMLData input ) : int
ClearContext ( ) : void	Clear any data from any context layers.
Clone ( ) : Object	Return a clone of this neural network. Including structure, weights and bias values. This is a deep copy.
Compute ( IMLData input ) : IMLData	Compute the output for a given input to the neural network.
Compute ( double input, double output ) : void	Compute the output for this network.
DecodeFromArray ( double encoded ) : void
DumpWeights ( ) : String
EnableConnection ( int fromLayer, int fromNeuron, int toNeuron, bool enable ) : void	Enable, or disable, a connection.
EncodeToArray ( double encoded ) : void
EncodedArrayLength ( ) : int
Equals ( BasicNetwork other ) : bool	Compare the two neural networks. For them to be equal they must be of the same structure, and have the same matrix values.
Equals ( BasicNetwork other, int precision ) : bool	Determine if this neural network is equal to another. Equal neural networks have the same weight matrix and bias values, within a specified precision.
GetActivation ( int layer ) : IActivationFunction	Get the activation function for the specified layer.
GetHashCode ( ) : int	Generate a hash code.
GetLayerBiasActivation ( int l ) : double	Get the bias activation for the specified layer.
GetLayerNeuronCount ( int l ) : int	Get the neuron count.
GetLayerOutput ( int layer, int neuronNumber ) : double	Get the layer output for the specified neuron.
GetLayerTotalNeuronCount ( int l ) : int	Get the total (including bias and context) neuron cont for a layer.
GetWeight ( int fromLayer, int fromNeuron, int toNeuron ) : double	Get the weight between the two layers.
IsConnected ( int layer, int fromNeuron, int toNeuron ) : bool	Determine if the specified connection is enabled.
IsLayerBiased ( int l ) : bool	Determine if the specified layer is biased.
Reset ( ) : void	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.
Reset ( int seed ) : void	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. Use the specified seed.
SetLayerBiasActivation ( int l, double v ) : void	Set the bias activation for the specified layer.
SetWeight ( int fromLayer, int fromNeuron, int toNeuron, double v ) : void	Set the weight between the two specified neurons.
ToString ( ) : String
UpdateProperties ( ) : void
ValidateNeuron ( int targetLayer, int neuron ) : void	Validate the the specified targetLayer and neuron are valid.
Winner ( IMLData input ) : int	Determine the winner for the specified input. This is the number of the winning neuron.

Описание методов

AddLayer() публичный Метод

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.

public AddLayer ( ILayer layer ) : void
layer	ILayer	The layer to be added to the network.
Результат	void

AddWeight() публичный Метод

Add to a weight.

public AddWeight ( int fromLayer, int fromNeuron, int toNeuron, double v ) : void
fromLayer	int	The from layer.
fromNeuron	int	The from neuron.
toNeuron	int	The to neuron.
v	double	The value to add.
Результат	void

BasicNetwork() публичный Метод

Construct an empty neural network.

public BasicNetwork ( ) : System
Результат	System

CalculateError() публичный Метод

Calculate the error for this neural network.

public CalculateError ( IMLDataSet data ) : double
data	IMLDataSet	The training set.
Результат	double

CalculateNeuronCount() публичный Метод

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

public CalculateNeuronCount ( ) : int
Результат	int

Classify() публичный Метод

public Classify ( IMLData input ) : int
input	IMLData
Результат	int

ClearContext() публичный Метод

Clear any data from any context layers.

public ClearContext ( ) : void
Результат	void

Clone() публичный Метод

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

public Clone ( ) : Object
Результат	Object

Compute() публичный Метод

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

public Compute ( IMLData input ) : IMLData
input	IMLData	The input to the neural network.
Результат	IMLData

Compute() публичный Метод

Compute the output for this network.

public Compute ( double input, double output ) : void
input	double	The input.
output	double	The output.
Результат	void

DecodeFromArray() публичный Метод

public DecodeFromArray ( double encoded ) : void
encoded	double
Результат	void

DumpWeights() публичный Метод

public DumpWeights ( ) : String
Результат	String

EnableConnection() публичный Метод

Enable, or disable, a connection.

public EnableConnection ( int fromLayer, int fromNeuron, int toNeuron, bool enable ) : void
fromLayer	int	The layer that contains the from neuron.
fromNeuron	int	The source neuron.
toNeuron	int	The target connection.
enable	bool	True to enable, false to disable.
Результат	void

EncodeToArray() публичный Метод

public EncodeToArray ( double encoded ) : void
encoded	double
Результат	void

EncodedArrayLength() публичный Метод

public EncodedArrayLength ( ) : int
Результат	int

Equals() публичный Метод

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

public Equals ( BasicNetwork other ) : bool
other	BasicNetwork	The other neural network.
Результат	bool

Equals() публичный Метод

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

public Equals ( BasicNetwork other, int precision ) : bool
other	BasicNetwork	The other neural network.
precision	int	The number of decimal places to compare to.
Результат	bool

GetActivation() публичный Метод

Get the activation function for the specified layer.

public GetActivation ( int layer ) : IActivationFunction
layer	int	The layer.
Результат	IActivationFunction

GetHashCode() публичный закрытый Метод

Generate a hash code.

public final GetHashCode ( ) : int
Результат	int

GetLayerBiasActivation() публичный Метод

Get the bias activation for the specified layer.

public GetLayerBiasActivation ( int l ) : double
l	int	The layer.
Результат	double

GetLayerNeuronCount() публичный Метод

Get the neuron count.

public GetLayerNeuronCount ( int l ) : int
l	int	The layer.
Результат	int

GetLayerOutput() публичный Метод

Get the layer output for the specified neuron.

public GetLayerOutput ( int layer, int neuronNumber ) : double
layer	int	The layer.
neuronNumber	int	The neuron number.
Результат	double

GetLayerTotalNeuronCount() публичный Метод

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

public GetLayerTotalNeuronCount ( int l ) : int
l	int	The layer.
Результат	int

GetWeight() публичный Метод

Get the weight between the two layers.

public GetWeight ( int fromLayer, int fromNeuron, int toNeuron ) : double
fromLayer	int	The from layer.
fromNeuron	int	The from neuron.
toNeuron	int	The to neuron.
Результат	double

IsConnected() публичный Метод

Determine if the specified connection is enabled.

public IsConnected ( int layer, int fromNeuron, int toNeuron ) : bool
layer	int	The layer to check.
fromNeuron	int	The source neuron.
toNeuron	int	THe target neuron.
Результат	bool

IsLayerBiased() публичный Метод

Determine if the specified layer is biased.

public IsLayerBiased ( int l ) : bool
l	int	The layer number.
Результат	bool

Reset() публичный Метод

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.

public Reset ( ) : void
Результат	void

Reset() публичный Метод

public Reset ( int seed ) : void
seed	int
Результат	void

SetLayerBiasActivation() публичный Метод

Set the bias activation for the specified layer.

public SetLayerBiasActivation ( int l, double v ) : void
l	int	The layer to use.
v	double
Результат	void

SetWeight() публичный Метод

Set the weight between the two specified neurons.

public SetWeight ( int fromLayer, int fromNeuron, int toNeuron, double v ) : void
fromLayer	int	The from layer.
fromNeuron	int	The from neuron.
toNeuron	int	The to neuron.
v	double	The to value.
Результат	void

ToString() публичный закрытый Метод

public final ToString ( ) : String
Результат	String

UpdateProperties() публичный закрытый Метод

public final UpdateProperties ( ) : void
Результат	void

ValidateNeuron() публичный Метод

Validate the the specified targetLayer and neuron are valid.

public ValidateNeuron ( int targetLayer, int neuron ) : void
targetLayer	int	The target layer.
neuron	int	The target neuron.
Результат	void

Winner() публичный Метод

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

public Winner ( IMLData input ) : int
input	IMLData	The input patter to present to the neural network.
Результат	int