Class CostFunction

Represents a cost function with vector arguments.

Inheritance

Object

CostFunction

CrossEntropyCost

HuberCost

MSECost

Inherited Members

Object.Equals(Object)

Object.Equals(Object, Object)

Object.GetHashCode()

Object.GetType()

Object.MemberwiseClone()

Object.ReferenceEquals(Object, Object)

Object.ToString()

Namespace: NeuralNetLearning.Maths.CostFunctions

Assembly: NeuralNet.dll

Syntax

public abstract class CostFunction

Methods

Apply(Vector<Double>, Vector<Double>)

Returns the numeric cost of the calculated vector predicted vs the target vector expected.

Declaration

public abstract double Apply(Vector<double> predicted, Vector<double> expected)

Parameters

Type	Name	Description
MathNet.Numerics.LinearAlgebra.Vector<Double>	predicted
MathNet.Numerics.LinearAlgebra.Vector<Double>	expected

Returns

Type	Description
Double

Derivative(Vector<Double>, Vector<Double>)

Returns the vector of derivatives of the cost function with respect to the calculated vector predicted.

Declaration

public abstract Vector<double> Derivative(Vector<double> predicted, Vector<double> expected)

Parameters

Type	Name	Description
MathNet.Numerics.LinearAlgebra.Vector<Double>	predicted
MathNet.Numerics.LinearAlgebra.Vector<Double>	expected

Returns

Type	Description
MathNet.Numerics.LinearAlgebra.Vector<Double>