Class ANN_MLP

Namespace

Emgu.CV.ML

Assembly

Emgu.CV.dll

Neural network

public class ANN_MLP : UnmanagedObject, IDisposable, IStatModel, IAlgorithm

Inheritance

object

DisposableObject

UnmanagedObject

ANN_MLP

Implements

IDisposable

IStatModel

IAlgorithm

Inherited Members

UnmanagedObject._ptr

UnmanagedObject.Ptr

DisposableObject.Dispose()

DisposableObject.ReleaseManagedResources()

object.GetType()

object.MemberwiseClone()

object.ToString()

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

Extension Methods

AlgorithmExtensions.Clear(IAlgorithm)

AlgorithmExtensions.GetDefaultName(IAlgorithm)

AlgorithmExtensions.IsEmpty(IAlgorithm)

AlgorithmExtensions.Load(IAlgorithm, string, string, string)

AlgorithmExtensions.LoadFromString(IAlgorithm, string, string, string)

AlgorithmExtensions.Read(IAlgorithm, FileNode)

AlgorithmExtensions.Save(IAlgorithm, string)

AlgorithmExtensions.SaveToString(IAlgorithm, string)

AlgorithmExtensions.Write(IAlgorithm, FileStorage, string)

AlgorithmExtensions.Write(IAlgorithm, FileStorage)

StatModelExtensions.Predict(IStatModel, IInputArray, IOutputArray, int)

StatModelExtensions.Train(IStatModel, IInputArray, DataLayoutType, IInputArray)

StatModelExtensions.Train(IStatModel, TrainData, int)

Constructors

ANN_MLP()

Create a neural network using the specific parameters

public ANN_MLP()

Properties

AnnealCoolingRatio

ANNEAL: Update cooling ratio.

public double AnnealCoolingRatio { get; set; }

Property Value

double

AnnealFinalT

ANNEAL: Update final temperature.

public double AnnealFinalT { get; set; }

Property Value

double

AnnealInitialT

ANNEAL: Update initial temperature.

public double AnnealInitialT { get; set; }

Property Value

double

AnnealItePerStep

ANNEAL: Update iteration per step.

public int AnnealItePerStep { get; set; }

Property Value

int

BackpropMomentumScale

BPROP: Strength of the momentum term (the difference between weights on the 2 previous iterations)

public double BackpropMomentumScale { get; set; }

Property Value

double

BackpropWeightScale

BPROP: Strength of the weight gradient term

public double BackpropWeightScale { get; set; }

Property Value

double

RpropDW0

RPROP: Initial value Delta_0 of update-values Delta_{ij}

public double RpropDW0 { get; set; }

Property Value

double

RpropDWMax

RPROP: Update-values upper limit

public double RpropDWMax { get; set; }

Property Value

double

RpropDWMin

RPROP: Update-values lower limit

public double RpropDWMin { get; set; }

Property Value

double

RpropDWMinus

RPROP: Decrease factor

public double RpropDWMinus { get; set; }

Property Value

double

RpropDWPlus

RPROP: Increase factor

public double RpropDWPlus { get; set; }

Property Value

double

TermCriteria

Termination criteria of the training algorithm

public MCvTermCriteria TermCriteria { get; set; }

Property Value

MCvTermCriteria

Methods

DisposeObject()

Release the memory associated with this neural network

protected override void DisposeObject()

SetActivationFunction(AnnMlpActivationFunction, double, double)

Initialize the activation function for each neuron.

public void SetActivationFunction(ANN_MLP.AnnMlpActivationFunction function, double param1 = 0, double param2 = 0)

Parameters

function ANN_MLP.AnnMlpActivationFunction

Currently the default and the only fully supported activation function is SigmoidSym

param1 double

The first parameter of the activation function.

param2 double

The second parameter of the activation function.

SetLayerSizes(IInputArray)

Sets the layer sizes.

public void SetLayerSizes(IInputArray layerSizes)

Parameters

layerSizes IInputArray

Integer vector specifying the number of neurons in each layer including the input and output layers. The very first element specifies the number of elements in the input layer. The last element - number of elements in the output layer.

SetTrainMethod(AnnMlpTrainMethod, double, double)

Sets training method and common parameters.

public void SetTrainMethod(ANN_MLP.AnnMlpTrainMethod method = AnnMlpTrainMethod.Rprop, double param1 = 0, double param2 = 0)

Parameters

method ANN_MLP.AnnMlpTrainMethod

The training method.

param1 double

param1 passed to setRpropDW0 for ANN_MLP::RPROP and to setBackpropWeightScale for ANN_MLP::BACKPROP and to initialT for ANN_MLP::ANNEAL.

param2 double

param2 passed to setRpropDWMin for ANN_MLP::RPROP and to setBackpropMomentumScale for ANN_MLP::BACKPROP and to finalT for ANN_MLP::ANNEAL.