C# Class Encog.Neural.Networks.Layers.RadialBasisFunctionLayer

Inheritance: BasicLayer
Exibir arquivo Open project: encog/encog-silverlight-core Class Usage Examples

Public Methods

Method Description
Compute ( INeuralData pattern ) : INeuralData

Compute the values before sending output to the next layer. This function allows the activation functions to be called.

CreatePersistor ( ) : IPersistor

Create a persistor for this layer.

RadialBasisFunctionLayer ( ) : System

Default constructor, mainly so the workbench can easily create a default layer.

RadialBasisFunctionLayer ( int neuronCount ) : System

Construct a radial basis function layer.

RandomizeRBFCentersAndWidths ( int dimensions, double min, double max, RBFEnum t ) : void

Set the gausian components to random values.

SaveOutNeuronCentersAndWeights ( double centers, double widths ) : void
SetRBFCentersAndWidths ( double centers, double widths, RBFEnum RBFType ) : void

Specify specific centers and widths for the provided RBFType

SetRBFCentersAndWidthsEqualSpacing ( double minPosition, double maxPosition, RBFEnum RBFType, int dimensions, double volumeNeuronRBFWidth, bool useWideEdgeRBFs ) : void

Equally spaces all hidden neurons within the n dimensional variable space.

Private Methods

Method Description
SetRBFFunction ( int RBFIndex, RBFEnum RBFType, double centers, double width ) : void

Method Details

Compute() public method

Compute the values before sending output to the next layer. This function allows the activation functions to be called.
public Compute ( INeuralData pattern ) : INeuralData
pattern INeuralData The incoming Project.
return INeuralData

CreatePersistor() public method

Create a persistor for this layer.
public CreatePersistor ( ) : IPersistor
return IPersistor

RadialBasisFunctionLayer() public method

Default constructor, mainly so the workbench can easily create a default layer.
public RadialBasisFunctionLayer ( ) : System
return System

RadialBasisFunctionLayer() public method

Construct a radial basis function layer.
public RadialBasisFunctionLayer ( int neuronCount ) : System
neuronCount int The neuron count.
return System

RandomizeRBFCentersAndWidths() public method

Set the gausian components to random values.
public RandomizeRBFCentersAndWidths ( int dimensions, double min, double max, RBFEnum t ) : void
dimensions int The number of dimensions in the network.
min double The minimum value for the centers, widths and peaks.
max double The maximum value for the centers, widths and peaks.
t RBFEnum The RBF to use.
return void

SaveOutNeuronCentersAndWeights() public method

public SaveOutNeuronCentersAndWeights ( double centers, double widths ) : void
centers double
widths double
return void

SetRBFCentersAndWidths() public method

Specify specific centers and widths for the provided RBFType
public SetRBFCentersAndWidths ( double centers, double widths, RBFEnum RBFType ) : void
centers double Array containing center position. Row n contains centers for neuron n. Row n contains x elements for x number of dimensions.
widths double Array containing widths. Row n contains widths for neuron n. Row n contains x elements for x number of dimensions.
RBFType RBFEnum The RBF Function to use for this layer.
return void

SetRBFCentersAndWidthsEqualSpacing() public method

Equally spaces all hidden neurons within the n dimensional variable space.
public SetRBFCentersAndWidthsEqualSpacing ( double minPosition, double maxPosition, RBFEnum RBFType, int dimensions, double volumeNeuronRBFWidth, bool useWideEdgeRBFs ) : void
minPosition double The minimum position neurons should be centered. Typically 0.
maxPosition double The maximum position neurons should be centered. Typically 1
RBFType RBFEnum The RBF type to use.
dimensions int The number of dimensions.
volumeNeuronRBFWidth double The neuron width of neurons within the mesh.
useWideEdgeRBFs bool Enables wider RBF's around the boundary of the neuron mesh.
return void