C# Класс Accord.Neuro.Learning.DeepNeuralNetworkLearning

Deep Neural Network learning algorithm.

Наследование: ISupervisedLearning

Показать файл Открыть проект Примеры использования класса

Открытые методы

Метод	Описание
ComputeError ( double inputs, double outputs ) : double	Computes the reconstruction error for a given set of input values.
DeepNeuralNetworkLearning ( DeepBeliefNetwork network ) : System	Creates a new DeepBeliefNetworkLearning algorithm.
GetLayerInput ( double input ) : double[][]	Gets the learning data needed to train the currently selected layer. The return of this function should then be passed to RunEpoch(double[][], double[][]) to actually run a learning epoch.
GetLayerInput ( double batches ) : double[][][]	Gets the learning data needed to train the currently selected layer. The return of this function should then be passed to RunEpoch(double[][], double[][]) to actually run a learning epoch.
Run ( double input, double output ) : double	Runs a single learning iteration.
RunEpoch ( double input, double output ) : double	Runs a single batch epoch of the learning algorithm.

Приватные методы

Метод	Описание
createAlgorithms ( ) : void

Описание методов

ComputeError() публичный Метод

Computes the reconstruction error for a given set of input values.

public ComputeError ( double inputs, double outputs ) : double
inputs	double	The input values.
outputs	double	The corresponding output values.
Результат	double

DeepNeuralNetworkLearning() публичный Метод

Creates a new DeepBeliefNetworkLearning algorithm.

public DeepNeuralNetworkLearning ( DeepBeliefNetwork network ) : System
network	Accord.Neuro.Networks.DeepBeliefNetwork	The network to be trained.
Результат	System

GetLayerInput() публичный Метод

Gets the learning data needed to train the currently selected layer. The return of this function should then be passed to RunEpoch(double[][], double[][]) to actually run a learning epoch.

public GetLayerInput ( double input ) : double[][]
input	double	The batch of input data.
Результат	double[][]

GetLayerInput() публичный Метод

Gets the learning data needed to train the currently selected layer. The return of this function should then be passed to RunEpoch(double[][], double[][]) to actually run a learning epoch.

public GetLayerInput ( double batches ) : double[][][]
batches	double	The mini-batches of input data.
Результат	double[][][]

Run() публичный Метод

Runs a single learning iteration.

public Run ( double input, double output ) : double
input	double	A single input vector.
output	double	The corresponding output vector.
Результат	double

RunEpoch() публичный Метод

Runs a single batch epoch of the learning algorithm.

public RunEpoch ( double input, double output ) : double
input	double	Array of input vectors.
output	double	Array of corresponding output vectors.
Результат	double