NeuralNetwork

Assembly: NN;

Namespace: Prognoz.Platform.Interop.NN;

Description

The NeuralNetwork class is used to work with artificial neural networks.

Comments

Using this class the user can create, learn and use the back-propagation network or self-organizing Kohonen map.

Class object methods inherited from INeuralNetwork

	Method name	Brief description
	ApplyConvexCombinationFactor	The ApplyConvexCombinationFactor method applies convex combinatorial transformation to the input elements of the network.
	CreateNetwork	The CreateNetwork method creates a neural network according to the assigned string view.
	CreateNetworkEx	The CreateNetworkEx method creates a neural network according to the assigned parameters.
	DeleteNetwork	The DeleteNetwork method deletes neural network.
	DeltasMinimumReachedBP	The DeltasMinimumReachedBP method returns whether the delta value has attained the assigned level.
	ExportSynapses	The ExportSynapses method returns string view of the network.
	GetClosestNeuron	The GetClosestNeuron method returns neuron index, which weight vector being least of all different from the tested input vector.
	GetError	The GetError method returns code of the last error of the neural network.
	GetInputValues	The GetInputValues method returns real array of the network input values.
	GetLearnRadius	The GetLearnRadius method returns value of the learning step for the specified layer.
	GetLearnRate	The GetLearnRate method returns learning rate value for the specified layer.
	GetMaximumWeightDelta	The GetMaximumWeightDelta method returns the maximum delta value of the weight of all the synapses.
	GetNumberOfInputs	The GetNumberOfInputs method returns the number of network inputs.
	GetNumberOfLayers	The GetNumberOfLayers method returns the number of neural network layers.
	GetNumberOfOutputs	The GetNumberOfOutputs method returns the number of network outputs.
	GetOutputValues	The GetOutputValues method returns the array of network output values.
	GetOutputWidth	The GetOutputWidth method returns the number of the lines in the output layer of the Kohonen self-organizing map.
	GetRowWidth	The GetRowWidth method returns the number of the lines in the output layer of the Kohonen self-organizing map.
	GetRowWidthEx	The GetRowWidthEx method returns the number of the lines in the specified layer of the Kohonen self-organizing map.
	GetSynapse	The GetSynapse method returns the weight value of the specified synapse.
	GetUseVectorScalar	The GetUseVectorScalar method returns whether the algorithm of scalar multiplication of vectors is used to calculate the distance between neurons in the specified layer.
	ImportSynapses	The ImportSynapses method loads values of synapse weights from the string view.
	InitSynapses	The InitSynapses method assigns values of synapse weights for the specified layer according to the assigned parameters.
	InitSynapsesConvex	The InitSynapsesConvex method assigns values of synapse weights by using the convex combination algorithm.
	InitSynapsesConvexEx	The InitSynapsesConvexEx method assigns values of synapse weights for the specified layer by using the convex combination algorithm.
	LearnBack	The LearnBack method performs iteration on learning back-propagation network.
	LearnSOFM	The LearnSOFM method learns a Kohonen self-organizing map.
	NormalizeInputValues	The NormalizeInputValues method normalizes network input data.
	NormalizeInputValuesEx	The NormalizeInputValuesEx method normalizes input values of the specified layer.
	NormalizeSynapses	The NormalizeSynapses method normalizes values of synapse weights of all the network layers.
	NormalizeSynapsesEx	The NormalizeSynapsesEx method normalizes values of synapse weights of the specified layer.
	PropagateBP	The PropagateBP method propagates the signal in the back-propagation network.
	PropagateSOFM	The PropagateSOFM method propagates the signal on the Kohonen self-organizing network.
	SetInputValues	The SetInputValues method sets network input values.
	SetInputValuesConvex	The SetInputValuesConvex method sets network input values by using the convex combinatorial transformation.
	SetInputValuesConvexEx	The SetInputValuesConvexEx method sets network input values by using the convex combinatorial transformation and ability of normalization.
	SetLearnRadius	The SetLearnRadius method sets network learning step.
	SetLearnRadiusEx	The SetLearnRadiusEx method sets network learning step.
	SetLearnRate	The SetLearnRate method sets learning rate.
	SetLearnRateEx	The SetLearnRateEx method sets leaning rate for the specified layer.
	SetMju	The SetMju method sets coefficient of inertia for learning of all the layers of the back-propagation network.
	SetMjuEx	The SetMjuEx method sets coefficient of inertia of learning for the specified layer of the back-propagation network.
	SetNju	The SetNju method sets learning speed of the back-propagation network.
	SetNjuEx	The SetNjuEx method sets learning speed for the specified layer of the back-propagation network.
	SetOutputWidth	The SetOutputWidth method sets the number of the lines in the output layer of the Kohonen self-organizing map.
	SetRowWidth	The SetRowWidth method sets the number of the lines in the output layer of the Kohonen self-organizing map.
	SetRowWidthEx	The SetRowWidthEx method sets the number of the lines in the specified layer of the Kohonen self-organizing map.
	SetSigmoidAlpha	The SetSigmoidAlpha method sets value of the Alpha coefficient for the sigmoid functions of signal propagation in the network.
	SetSigmoidFuncs	The SetSigmoidFuncs method sets the type of signal propagation in the network.
	SetSigmoidFuncsEx	The SetSigmoidFuncs method sets the type of signal propagation in the network for the specified layer.
	SetSynapse	The SetSynapse method sets the value of the specified synapse weight.
	SetUseVectorScalar	The SetUseVectorScalar method determines whether to apply the algorithm of scalar multiplication of vectors to calculate the distance between neurons.
	SetUseVectorScalarEx	The SetUseVectorScalarEx method determines whether to apply the algorithm of scalar multiplication of vectors to calculate the distance between neurons of the specified layer.

See also:

NN Assembly Classes