Interface IsotropicKernel

All Superinterfaces:
Function, Serializable
All Known Implementing Classes:
BinarySparseGaussianKernel, BinarySparseLaplacianKernel, BinarySparseMaternKernel, BinarySparseThinPlateSplineKernel, Gaussian, GaussianKernel, Laplacian, LaplacianKernel, Matern, MaternKernel, SparseGaussianKernel, SparseLaplacianKernel, SparseMaternKernel, SparseThinPlateSplineKernel, ThinPlateSpline, ThinPlateSplineKernel
public interface IsotropicKernel extends Function
Isotropic kernel. If the kernel is a function only of the distance |x − y|, it is called isotropic. It is thus invariant to all rigid motions.

  • Field Summary

    Fields inherited from interface Function

    logger

  • Method Summary

    Modifier and Type
    Method
    Description
    default double
    apply(double dist)
    Computes the kernel function.
    default double
    f(double dist)
    Computes the value of the function at x.
    double
    k(double dist)
    Computes the isotropic kernel function.
    default DenseMatrix
    K(DenseMatrix pdist)
    Computes the kernel matrix.
    double[]
    kg(double dist)
    Computes the isotropic kernel function and its gradient over hyperparameters.
    default DenseMatrix[]
    KG(DenseMatrix pdist)
    Computes the kernel and gradient matrices.

    Methods inherited from interface Function

    inv, root

  • Method Details

    • f

      default double f(double dist)
      Description copied from interface: Function
      Computes the value of the function at x.
      Specified by:
      f in interface Function
      Parameters:
      dist - a real number.
      Returns:
      the function value.

    • k

      double k(double dist)
      Computes the isotropic kernel function.
      Parameters:
      dist - The distance.
      Returns:
      the kernel value.

    • kg

      double[] kg(double dist)
      Computes the isotropic kernel function and its gradient over hyperparameters.
      Parameters:
      dist - The distance.
      Returns:
      the kernel value and gradient.

    • apply

      default double apply(double dist)
      Computes the kernel function. This is simply for Scala convenience.
      Specified by:
      apply in interface Function
      Parameters:
      dist - The distance.
      Returns:
      the kernel value.

    • K

      default DenseMatrix K(DenseMatrix pdist)
      Computes the kernel matrix.
      Parameters:
      pdist - The pairwise distance matrix.
      Returns:
      The kernel matrix.

    • KG

      default DenseMatrix[] KG(DenseMatrix pdist)
      Computes the kernel and gradient matrices.
      Parameters:
      pdist - The pairwise distance matrix.
      Returns:
      the kernel and gradient matrices.