kmeans::Initialize< Index_, Data_, Cluster_, Float_, Matrix_ > Class Template Reference

Interface for k-means initialization algorithms. More...

#include <Initialize.hpp>

Inheritance diagram for kmeans::Initialize< Index_, Data_, Cluster_, Float_, Matrix_ >:

Public Member Functions
virtual Cluster_	run (const Matrix_ &data, Cluster_ num_centers, Float_ *centers) const =0

Detailed Description

template<typename Index_, typename Data_, typename Cluster_, typename Float_, class Matrix_ = Matrix<Data_, Index_>>
class kmeans::Initialize< Index_, Data_, Cluster_, Float_, Matrix_ >

Interface for k-means initialization algorithms.

Template Parameters

Index_	Integer type for the observation indices in the input dataset.
Data_	Numeric type for the input dataset.
Cluster_	Integer type for the cluster assignments.
Float_	Floating-point type for the centroids. This will also be used for any internal distance calculations.
Matrix_	Class of the input data matrix. This should satisfy the Matrix interface.

Member Function Documentation

run()

template<typename Index_ , typename Data_ , typename Cluster_ , typename Float_ , class Matrix_ = Matrix<Data_, Index_>>

virtual Cluster_ kmeans::Initialize< Index_, Data_, Cluster_, Float_, Matrix_ >::run ( const Matrix_ & data, Cluster_ num_centers, Float_ * centers ) const

pure virtual

Parameters

	data	A matrix-like object containing per-observation data.
	num_centers	Number of cluster centers.
[out]	centers	Pointer to an array of length equal to the product of num_centers and data.num_dimensions(). This contains a column-major matrix where rows correspond to dimensions and columns correspond to cluster centers. On output, each column will contain the final centroid locations for each cluster.

Returns

centers is filled with the new cluster centers. The number of filled centers is returned - this is usually equal to num_centers, but may not be if, e.g., num_centers is greater than the number of observations. If the returned value is less than num_centers, only the first few centers in centers will be filled.

---

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

• kmeans/Initialize.hpp