cmp::kernel::DiagonalBandwidth Class Reference

Diagonal bandwidth matrix (independent parameter per dimension). More...

#include <kernel++.h>

Inheritance diagram for cmp::kernel::DiagonalBandwidth:

Collaboration diagram for cmp::kernel::DiagonalBandwidth:

Public Member Functions
	DiagonalBandwidth ()=delete
	DiagonalBandwidth (const Eigen::VectorXd &h)
	Constructs a DiagonalBandwidth from a parameter vector.
Eigen::VectorXd	apply (const Eigen::VectorXd &x1, const Eigen::VectorXd &x2) const override
	Scales the coordinate distances independently by 1 / h_d.
double	determinant () const override
	Computes the determinant of the diagonal bandwidth matrix.
size_t	size () const override
	Returns the dimension of the space.
Eigen::VectorXd	gradientOfApply (const Eigen::VectorXd &x1, const Eigen::VectorXd &x2, const size_t &i) const override
double	gradientOfLogDeterminant (const size_t &i) const override
void	setFromVector (const Eigen::VectorXd &params) override
Eigen::VectorXd	getParams () const override
Public Member Functions inherited from cmp::kernel::Bandwidth
virtual	~Bandwidth ()=default

Static Public Member Functions
static std::shared_ptr< Bandwidth >	make (const Eigen::VectorXd &h)

Private Attributes
Eigen::VectorXd	h_
	Diagonal bandwidth parameters vector.
double	det_
	Precomputed determinant of the diagonal bandwidth matrix.

Detailed Description

Diagonal bandwidth matrix (independent parameter per dimension).

Mathematical Formulation Represents a diagonal matrix \(\mathbf{H} = \text{diag}(h_1, \dots, h_D)\). Coordinates are scaled independently:

\[ z_d = \frac{x_{1,d} - x_{2,d}}{h_d} \]

The determinant is \(\det(\mathbf{H}) = \prod_{d=1}^D h_d\).

Constructor & Destructor Documentation

DiagonalBandwidth() [1/2]

cmp::kernel::DiagonalBandwidth::DiagonalBandwidth ( )

delete

DiagonalBandwidth() [2/2]

cmp::kernel::DiagonalBandwidth::DiagonalBandwidth ( const Eigen::VectorXd &  h )

inline

Constructs a DiagonalBandwidth from a parameter vector.

Member Function Documentation

apply()

Eigen::VectorXd cmp::kernel::DiagonalBandwidth::apply ( const Eigen::VectorXd &  x1, const Eigen::VectorXd &  x2  ) const

overridevirtual

Scales the coordinate distances independently by 1 / h_d.

Implements cmp::kernel::Bandwidth.

determinant()

double cmp::kernel::DiagonalBandwidth::determinant ( ) const

overridevirtual

Computes the determinant of the diagonal bandwidth matrix.

Implements cmp::kernel::Bandwidth.

getParams()

Eigen::VectorXd cmp::kernel::DiagonalBandwidth::getParams ( ) const

overridevirtual

Implements cmp::kernel::Bandwidth.

gradientOfApply()

Eigen::VectorXd cmp::kernel::DiagonalBandwidth::gradientOfApply ( const Eigen::VectorXd &  x1, const Eigen::VectorXd &  x2, const size_t &  i  ) const

overridevirtual

Implements cmp::kernel::Bandwidth.

gradientOfLogDeterminant()

double cmp::kernel::DiagonalBandwidth::gradientOfLogDeterminant ( const size_t &  i ) const

overridevirtual

Implements cmp::kernel::Bandwidth.

make()

static std::shared_ptr< Bandwidth > cmp::kernel::DiagonalBandwidth::make ( const Eigen::VectorXd &  h )

inlinestatic

setFromVector()

void cmp::kernel::DiagonalBandwidth::setFromVector ( const Eigen::VectorXd &  params )

overridevirtual

Implements cmp::kernel::Bandwidth.

size()

size_t cmp::kernel::DiagonalBandwidth::size ( ) const

inlineoverridevirtual

Returns the dimension of the space.

Implements cmp::kernel::Bandwidth.

Member Data Documentation

det_

double cmp::kernel::DiagonalBandwidth::det_

private

Precomputed determinant of the diagonal bandwidth matrix.

h_

Eigen::VectorXd cmp::kernel::DiagonalBandwidth::h_

private

Diagonal bandwidth parameters vector.

---

The documentation for this class was generated from the following files:

• include/kernel++.h
• src/kernel++.cpp