rtflann::GroupWiseCenterChooser< Distance > Class Template Reference

#include <center_chooser.h>

Inheritance diagram for rtflann::GroupWiseCenterChooser< Distance >:

Public Types
typedef Distance::ResultType	DistanceType
typedef Distance::ElementType	ElementType
Public Types inherited from rtflann::CenterChooser< Distance >
typedef Distance::ResultType	DistanceType
typedef Distance::ElementType	ElementType

Public Member Functions
	GroupWiseCenterChooser (const Distance &distance, const std::vector< ElementType * > &points)
void	operator() (int k, int *indices, int indices_length, int *centers, int &centers_length)
Public Member Functions inherited from rtflann::CenterChooser< Distance >
	CenterChooser (const Distance &distance, const std::vector< ElementType * > &points)
void	setDataSize (size_t cols)
virtual	~CenterChooser ()

Additional Inherited Members
Protected Attributes inherited from rtflann::CenterChooser< Distance >
size_t	cols_
const Distance	distance_
const std::vector< ElementType * > &	points_

Detailed Description

template<typename Distance>
class rtflann::GroupWiseCenterChooser< Distance >

Chooses the initial centers in a way inspired by Gonzales (by Pierre-Emmanuel Viel): select the first point of the list as a candidate, then parse the points list. If another point is further than current candidate from the other centers, test if it is a good center of a local aggregation. If it is, replace current candidate by this point. And so on...

Used with KMeansIndex that computes centers coordinates by averaging positions of clusters points, this doesn't make a real difference with previous methods. But used with HierarchicalClusteringIndex class that pick centers among existing points instead of computing the barycenters, there is a real improvement.

Definition at line 307 of file center_chooser.h.

Member Typedef Documentation

template<typename Distance>

typedef Distance::ResultType rtflann::GroupWiseCenterChooser< Distance >::DistanceType

Definition at line 311 of file center_chooser.h.

template<typename Distance>

typedef Distance::ElementType rtflann::GroupWiseCenterChooser< Distance >::ElementType

Definition at line 310 of file center_chooser.h.

Constructor & Destructor Documentation

template<typename Distance>

rtflann::GroupWiseCenterChooser< Distance >::GroupWiseCenterChooser ( const Distance &  distance, const std::vector< ElementType * > &  points  )

inline

Definition at line 317 of file center_chooser.h.

Member Function Documentation

template<typename Distance>

void rtflann::GroupWiseCenterChooser< Distance >::operator() ( int  k, int *  indices, int  indices_length, int *  centers, int &  centers_length  )

inlinevirtual

Chooses cluster centers

Parameters

k	number of centers to choose
indices	indices of points to choose the centers from
indices_length	length of indices
centers	indices of chosen centers
centers_length	length of centers array

Implements rtflann::CenterChooser< Distance >.

Definition at line 320 of file center_chooser.h.

---

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

• center_chooser.h