theplu::yat::regression::Local Class Reference

Class for Locally weighted regression. More...

#include <yat/regression/Local.h>

List of all members.

Public Member Functions
	Local (OneDimensionalWeighted &r, Kernel &k)
	Constructor taking type of regressor, type of kernel.
virtual	~Local (void)
	The destructor.
void	add (const double x, const double y)
void	fit (const size_t step_size, const size_t nof_points)
void	reset (void)
	Set everything to zero.
const utility::Vector &	x (void) const
const utility::Vector &	y_predicted (void) const
const utility::Vector &	y_err (void) const

---

Detailed Description

Class for Locally weighted regression.

Locally weighted regression is an algorithm for learning continuous non-linear mappings in a non-parametric manner. In locally weighted regression, points are weighted by proximity to the current x in question using a Kernel. A weighted regression is then computed using the weighted points and a specific Regression method. This procedure is repeated, which results in a pointwise approximation of the underlying (unknown) function.

---

Member Function Documentation

void theplu::yat::regression::Local::add	(	const double	x,
		const double	y
	)

adding a data point

void theplu::yat::regression::Local::fit	(	const size_t	step_size,
		const size_t	nof_points
	)

Parameters:

	step_size	Size of step between each fit
	nof_points	Number of points used in each fit

Exceptions:

std::runtime_error

if step_size is 0, nof_points is less than 3, or step_size is larger than number of added data points.

void theplu::yat::regression::Local::reset

(

void

)

Set everything to zero.

Since:

New in yat 0.5

const utility::Vector& theplu::yat::regression::Local::x

(

void

)

const

Returns:

x-values where fitting was performed.

const utility::Vector& theplu::yat::regression::Local::y_predicted

(

void

)

const

Function returning predicted values

const utility::Vector& theplu::yat::regression::Local::y_err

(

void

)

const

Function returning error of predictions

---

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

• yat/regression/Local.h