npstat::LogisticRegressionBase< Numeric > Class Template Reference

#include <LocalLogisticRegression.hh>

Inheritance diagram for npstat::LogisticRegressionBase< Numeric >:

Public Member Functions
	LogisticRegressionBase (const QuadraticOrthoPolyND &poly, bool calculateLikelihoodGradient)
unsigned	dim () const
const QuadraticOrthoPolyND &	getPoly () const
bool	calculatingGradient () const
const std::vector< double > &	lastCoeffs () const
void	getGradient (double *buffer, unsigned bufLen) const
double	getPassCount () const
double	getFailCount () const
void	setRegressionBox (const BoxND< Numeric > &box)
void	setLinearMapping (const double *location, const double *scale, unsigned locationAndScaleArraySize)
double	calculateLogLikelihood (const double *coeffs, unsigned nCoeffs)

Protected Member Functions
void	resetAccumulators ()

Protected Attributes
const QuadraticOrthoPolyND &	poly_
BoxND< Numeric >	regressionBox_
std::vector< double >	location_
std::vector< double >	scale_
std::vector< double >	coords_
std::vector< double >	coeffs_
std::vector< double >	gradBuffer_
std::vector< long double >	gradient_
long double	logli_
long double	passCount_
long double	failCount_
const double	minlog_
const double	maxlog_
const unsigned	mydim_
const bool	calcGradient_

Detailed Description

template<class Numeric>
class npstat::LogisticRegressionBase< Numeric >

Base class for logistic regression

Member Function Documentation

calculateLogLikelihood()

template<class Numeric >

double npstat::LogisticRegressionBase< Numeric >::calculateLogLikelihood	(	const double *	coeffs,
		unsigned	nCoeffs
	)

The next method calculates the negative log-likelihood of the local logistic regression fit for the current mapping. In case the corresponding flag was set "true" in the constructor, the gradient of this quantity will be calculated as well. The gradient can be subsequently retrieved with the "getGradient" method.

dim()

template<class Numeric >

unsigned npstat::LogisticRegressionBase< Numeric >::dim ( ) const

inline

Inspect object properties

lastCoeffs()

template<class Numeric >

const std::vector<double>& npstat::LogisticRegressionBase< Numeric >::lastCoeffs ( ) const

inline

Inspect results of the most recent fit

setLinearMapping()

template<class Numeric >

void npstat::LogisticRegressionBase< Numeric >::setLinearMapping	(	const double *	location,
		const double *	scale,
		unsigned	locationAndScaleArraySize
	)

The next method sets up a mapping from the fitted region into the region which was used to create the orthogonal polynomials. The mapping is linear in every coordinate separately: ortho[i] = fitted[i]*scale[i] + location[i]. Note that the "setRegressionBox" method automatically sets up the default mapping which will be adequate for most applications.

setRegressionBox()

template<class Numeric >

void npstat::LogisticRegressionBase< Numeric >::setRegressionBox

(

const BoxND< Numeric > &

box

)

Only the data inside the regression box will be used in the fit (and some points may be subsequently weighted down to zero by the weight function).

Note that the "setRegressionBox" method automatically sets up the standard mapping: the edges of the regression box are mapped into the boundaries which were used to create the orthogonal polynomials. If the standard mapping is not what you want (for example, this might be the case near the boundaries of the fitted region), you must call "setLinearMapping" after calling "setRegressionBox" in order to override the standard mapping.

This method must be called at least once after the object is constructed and before any call to "calculateLogLikelihood".

---

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

• stat/LocalLogisticRegression.hh