npstat::AbsSymbetaFilterProvider Class Reference

#include <AbsSymbetaFilterProvider.hh>

Inheritance diagram for npstat::AbsSymbetaFilterProvider:

Public Member Functions
virtual CPP11_shared_ptr< const LocalPolyFilter1D >	provideFilter (int symbetaPower, double bandwidth, double degree, unsigned nbins, double binwidth, const BoundaryHandling &bm, unsigned excludedBin, bool excludeCentralPoint)=0
virtual void	startMemoizing ()
virtual void	stopMemoizing ()
virtual bool	isMemoizing () const

Detailed Description

Interface definition for classes which build LOrPE filters using symmetric beta kernels (or a Gaussian kernel) as weights. The derived classes are supposed to override the "provideFilter" member function.

Member Function Documentation

isMemoizing()

virtual bool npstat::AbsSymbetaFilterProvider::isMemoizing ( ) const

inlinevirtual

Is the filter provider memoizing filters currently?

provideFilter()

virtual CPP11_shared_ptr<const LocalPolyFilter1D> npstat::AbsSymbetaFilterProvider::provideFilter ( int  symbetaPower, double  bandwidth, double  degree, unsigned  nbins, double  binwidth, const BoundaryHandling &  bm, unsigned  excludedBin, bool  excludeCentralPoint  )

pure virtual

Parameters are:

symbetaPower – This parameter defines which kernel to use. Negative value means use Gaussian.

bandwidth – Kernel bandwidth.

degree – Polynomial degree.

nbins – Number of bins in the discretized dataset.

binwidth – Bin width.

bm – Boundary handling method.

excludedBin – Bin number to exclude. Numbers equal to nbins and larger are to be ignored.

excludeCentralPoint – Exclude central bin from the weight function.

The function should return an instance of LocalPolyFilter1D wrapped into a smart pointer.

Implemented in npstat::MemoizingSymbetaFilterProvider, and npstat::SimpleSymbetaFilterProvider.

startMemoizing()

virtual void npstat::AbsSymbetaFilterProvider::startMemoizing ( )

inlinevirtual

This function can be used to advise the filter provider which filters should be stored for subsequent reuse. The filter provider is free to heed or to ignore this advice.

---

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

• stat/AbsSymbetaFilterProvider.hh