Search Results for

    Show / Hide Table of Contents

    Class KernelFunction

    Base class for all kernel functions

    Inheritance
    Object
    KernelFunction
    ARD
    LinearKernel
    NNKernel
    SquaredExponential
    SummationKernel
    WhiteNoise
    Implements
    IKernelFunctionWithParams
    IKernelFunction
    Inherited Members
    Object.Equals(Object)
    Object.Equals(Object, Object)
    Object.GetHashCode()
    Object.GetType()
    Object.MemberwiseClone()
    Object.ReferenceEquals(Object, Object)
    Object.ToString()
    Namespace: Microsoft.ML.Probabilistic.Distributions.Kernels
    Assembly: Microsoft.ML.Probabilistic.dll
    Syntax
    [Serializable]
    public abstract class KernelFunction : IKernelFunctionWithParams, IKernelFunction

    Constructors

    KernelFunction()

    Default constructor

    Declaration
    protected KernelFunction()

    KernelFunction(IList<String>)

    Protected constructor - derived classes pass down their list of hyper-parameter names

    Declaration
    protected KernelFunction(IList<string> hyperNames)
    Parameters
    Type Name Description
    IList<String> hyperNames

    KernelFunction(IList<String>, Double[])

    Declaration
    [Construction(new string[]{"thetaNames", "thetaValues"})]
    protected KernelFunction(IList<string> hyperNames, double[] values)
    Parameters
    Type Name Description
    IList<String> hyperNames
    Double[] values

    Fields

    thetaName2Index

    Dictionary that allows look-up of index from hype-parameter name

    Declaration
    protected Dictionary<string, int> thetaName2Index
    Field Value
    Type Description
    Dictionary<String, Int32>

    thetaNames

    Hyper-parameter names

    Declaration
    protected IList<string> thetaNames
    Field Value
    Type Description
    IList<String>

    thetaValues

    Hyper-parameter values

    Declaration
    protected double[] thetaValues
    Field Value
    Type Description
    Double[]

    Properties

    HyperNames

    Sets the names of the hyper-parameters. Note that this destroys any values

    Declaration
    public virtual IList<string> HyperNames { set; }
    Property Value
    Type Description
    IList<String>

    Item[Int32]

    Sets or gets a log hyper-parameter by index

    Declaration
    public virtual double this[int index] { get; set; }
    Parameters
    Type Name Description
    Int32 index

    Index of the log hyper-parameter

    Property Value
    Type Description
    Double

    The log hyper-parameter value

    Item[String]

    Sets or gets hyper-parameter by name. This indexer is not over-rideable

    Declaration
    public virtual double this[string name] { get; set; }
    Parameters
    Type Name Description
    String name

    Mame of the hyper-parameter

    Property Value
    Type Description
    Double

    The hyper-parameter value

    ThetaCount

    Hyper-parameter count

    Declaration
    public virtual int ThetaCount { get; }
    Property Value
    Type Description
    Int32

    TypeVersion

    The static version for the derived class

    Declaration
    public abstract int TypeVersion { get; }
    Property Value
    Type Description
    Int32

    Methods

    Cholesky(IKernelFunction, Dictionary<Int32, Vector>)

    Cholesky of Kernel matrix

    Declaration
    public static LowerTriangularMatrix Cholesky(IKernelFunction kf, Dictionary<int, Vector> xData)
    Parameters
    Type Name Description
    IKernelFunction kf

    Kernel function

    Dictionary<Int32, Vector> xData

    Data with which to build the matrix

    Returns
    Type Description
    LowerTriangularMatrix

    EvaluateX(Vector)

    Evaluates the kernel for a single vector

    Declaration
    public virtual double EvaluateX(Vector x)
    Parameters
    Type Name Description
    Vector x

    Vector

    Returns
    Type Description
    Double

    EvaluateX(Vector, ref Vector)

    Evaluates the kernel for a single vector, and optionally, returns the derivatives with respect to the parameters

    Declaration
    public virtual double EvaluateX(Vector x, ref Vector logThetaDeriv)
    Parameters
    Type Name Description
    Vector x

    Vector

    Vector logThetaDeriv

    Derivative of the kernel value with respect to the log hyper-parameters

    Returns
    Type Description
    Double

    EvaluateX(Vector, ref Vector, ref Vector)

    Evaluates the kernel for a single vector, and optionally, returns the derivatives with respect to the vector and with respect to the parameters

    Declaration
    public abstract double EvaluateX(Vector x, ref Vector xDeriv, ref Vector logThetaDeriv)
    Parameters
    Type Name Description
    Vector x

    Vector

    Vector xDeriv

    Derivative of the kernel value with respect to x

    Vector logThetaDeriv

    Derivative of the kernel value with respect to the log hyper-parameters

    Returns
    Type Description
    Double

    EvaluateX1X2(Vector, Vector)

    Evaluates the kernel for a pair of vectors

    Declaration
    public virtual double EvaluateX1X2(Vector x1, Vector x2)
    Parameters
    Type Name Description
    Vector x1

    First vector

    Vector x2

    Second vector

    Returns
    Type Description
    Double

    EvaluateX1X2(Vector, Vector, ref Vector)

    Evaluates the kernel for a pair of vectors and, optionally, returns derivatives with respect to the parameters

    Declaration
    public virtual double EvaluateX1X2(Vector x1, Vector x2, ref Vector logThetaDeriv)
    Parameters
    Type Name Description
    Vector x1

    First vector

    Vector x2

    Second vector

    Vector logThetaDeriv

    Derivative of the kernel value with respect to the log hyper-parameters

    Returns
    Type Description
    Double

    EvaluateX1X2(Vector, Vector, ref Vector, ref Vector)

    Evaluates the kernel for a pair of vectors and, optionally, returns derivatives with respect to each vector, and with respect to the parameters

    Declaration
    public abstract double EvaluateX1X2(Vector x1, Vector x2, ref Vector x1Deriv, ref Vector logThetaDeriv)
    Parameters
    Type Name Description
    Vector x1

    First vector

    Vector x2

    Second vector

    Vector x1Deriv

    Derivative of the kernel value with respect to x1 input vector

    Vector logThetaDeriv

    Derivative of the kernel value with respect to the log hyper-parameters

    Returns
    Type Description
    Double

    IndexToName(Int32)

    Gets the index of a specified hyperparameter name

    Declaration
    public virtual string IndexToName(int thetaIndex)
    Parameters
    Type Name Description
    Int32 thetaIndex

    The hyper-parameter index

    Returns
    Type Description
    String

    The index

    NameToIndex(String)

    Gets the index of a specified hyperparameter name

    Declaration
    public virtual int NameToIndex(string thetaName)
    Parameters
    Type Name Description
    String thetaName

    The hyper-parameter name

    Returns
    Type Description
    Int32

    The index

    Read(StreamReader)

    Read the parameters in from a stream

    Declaration
    public virtual void Read(StreamReader sr)
    Parameters
    Type Name Description
    StreamReader sr

    Stream reader

    Write(StreamWriter)

    Writes the parameters out to a stream

    Declaration
    public virtual void Write(StreamWriter sw)
    Parameters
    Type Name Description
    StreamWriter sw

    Stream writer

    Implements

    IKernelFunctionWithParams
    IKernelFunction
    In This Article
    Back to top Copyright © .NET Foundation. All rights reserved.