Interface DotProductKernel

All Superinterfaces:

Function, Serializable

All Known Implementing Classes:

BinarySparseHyperbolicTangentKernel, BinarySparseLinearKernel, BinarySparsePolynomialKernel, HyperbolicTangent, HyperbolicTangentKernel, LinearKernel, Polynomial, PolynomialKernel, SparseHyperbolicTangentKernel, SparseLinearKernel, SparsePolynomialKernel

public interface DotProductKernel
extends Function

Dot product kernel depends only on the dot product of x and y.

Field Summary

Fields inherited from interface Function

logger

Method Summary

Modifier and Type	Method	Description
default double	apply(double dot)	Computes the kernel function.
default double	f(double dot)	Computes the value of the function at x.
double	k(double dot)	Computes the dot product kernel function.
default Matrix	K(Matrix pdot)	Computes the kernel matrix.
double[]	kg(double dot)	Computes the dot product kernel function and its gradient over hyperparameters.

Methods inherited from interface Function

inv, root

Method Details

f

default double f(double dot)

Description copied from interface: Function

Computes the value of the function at x.

Specified by:

f in interface Function

Parameters:

dot - a real number.

Returns:

the function value.

k

double k(double dot)

Computes the dot product kernel function.

Parameters:

dot - the dot product.

Returns:

the kernel value.

kg

double[] kg(double dot)

Computes the dot product kernel function and its gradient over hyperparameters.

Parameters:

dot - The dot product.

Returns:

the kernel value and gradient.

apply

default double apply(double dot)

Computes the kernel function. This is simply for Scala convenience.

Specified by:

apply in interface Function

Parameters:

dot - the dot product.

Returns:

the kernel value.

K

default Matrix K(Matrix pdot)

Computes the kernel matrix.

Parameters:

pdot - the pairwise dot product matrix.

Returns:

the kernel matrix.