Class GWRScalable

Defined in File GWRScalable.h

Nested Relationships

Nested Types

Struct GWRScalable::LoocvParams

Inheritance Relationships

Base Type

public gwm::GWRBase (Class GWRBase)

Class Documentation

class GWRScalable : public gwm::GWRBase 

Scalable GWR.

Public Types

enum BandwidthSelectionCriterionType

Type of bandwidth criterion.

Values:

enumerator AIC: AIC.

enumerator CV: CV.

Public Functions

inline GWRScalable(): Construct a new GWRScalable object.

inline ~GWRScalable(): Destroy the GWRScalable object.

inline arma::uword polynomial() const: Get the degree of polynomial kernel The degree of polynomial kernel.

inline void setPolynomial(arma::uword polynomial): Set the degree of polynomial kernel The degree of polynomial kernel.

inline double cv() const: Get the value of CV criterion Value of CV criterion.

inline double scale() const: Get the scale. The value of scale.

inline double penalty() const: Get the penalty. The value of penalty.

inline bool hasHatMatrix() const: Get whether has hat matrix. Yes No.

inline void setHasHatMatrix(const bool has): Set whether has hat matrix. Whether has hat matrix.

inline BandwidthSelectionCriterionType parameterOptimizeCriterion() const: Get the type of calculator for parameter optimization criterion. Type of calculator for parameter optimization criterion.

inline void setParameterOptimizeCriterion(const BandwidthSelectionCriterionType &parameterOptimizeCriterion): Set the calculator for parameter optimization criterion. Type of calculator for parameter optimization criterion.

virtual bool isValid() override

Check whether the algorithm’s configuration is valid.

Returns:: true if the algorithm’s configuration is valid.
Returns:: false if the algorithm’s configuration is invalid.

virtual arma::mat fit() override

Fit coefficient estimates.

Returns:: mat Coefficient estimates

virtual arma::mat predict(const arma::mat &locations) override

Predict coefficients on specified locations.

Parameters:: locations – Locations where to predict coefficients.
Returns:: mat Predicted coefficients.

Public Static Functions

static double Loocv(const arma::vec &target, const arma::mat &x, const arma::vec &y, arma::uword poly, const arma::mat &Mx0, const arma::mat &My0): Calculate the value of CV criterion. Variables to optimize Independent variables Dependent variables The degree of polynomial kernel Value of CV criterion.

static double AICvalue(const arma::vec &target, const arma::mat &x, const arma::vec &y, arma::uword poly, const arma::mat &Mx0, const arma::mat &My0): Calculate the value of AIC criterion. Variables to optimize Independent variables Dependent variables The degree of polynomial kernel Value of AIC criterion.

struct LoocvParams

Additional parameters for leave-one-out CV.

Public Members

const arma::mat *x: Independent variables.

const arma::mat *y: Dependent variables.

const arma::uword polynomial: The degree of polynomial kernel.

const arma::mat *Mx0

const arma::mat *My0