kmeans.c File Reference

K-means - Declaration. More...

#include "kmeans.h"
#include "generic.h"
#include "mathop.h"
#include <string.h>
#include "shuffle-def.h"

Functions
void	vl_kmeans_reset (VlKMeans *self)
	Reset state. More...
VlKMeans *	vl_kmeans_new (vl_type dataType, VlVectorComparisonType distance)
	Create a new KMeans object. More...
VlKMeans *	vl_kmeans_new_copy (VlKMeans const *kmeans)
	Create a new KMeans object by copy. More...
void	vl_kmeans_delete (VlKMeans *self)
	Deletes a KMeans object. More...
void	vl_kmeans_set_centers (VlKMeans *self, void const *centers, vl_size dimension, vl_size numCenters)
	Set centers. More...
void	vl_kmeans_init_centers_with_rand_data (VlKMeans *self, void const *data, vl_size dimension, vl_size numData, vl_size numCenters)
	init centers by randomly sampling data More...
void	vl_kmeans_init_centers_plus_plus (VlKMeans *self, void const *data, vl_size dimension, vl_size numData, vl_size numCenters)
	Seed centers by the KMeans++ algorithm. More...
void	vl_kmeans_quantize (VlKMeans *self, vl_uint32 *assignments, void *distances, void const *data, vl_size numData)
	Quantize data. More...
void	vl_kmeans_quantize_ann (VlKMeans *self, vl_uint32 *assignments, void *distances, void const *data, vl_size numData, vl_bool update)
	Quantize data using approximate nearest neighbours (ANN). More...
double	vl_kmeans_refine_centers (VlKMeans *self, void const *data, vl_size numData)
	Refine center locations. More...
double	vl_kmeans_cluster (VlKMeans *self, void const *data, vl_size dimension, vl_size numData, vl_size numCenters)
	Cluster data. More...

Detailed Description

Author

Andrea Vedaldi, David Novotny

Function Documentation

vl_kmeans_cluster()

double vl_kmeans_cluster	(	VlKMeans *	self,
		void const *	data,
		vl_size	dimension,
		vl_size	numData,
		vl_size	numCenters
	)

Parameters

self	KMeans object.
data	data to quantize.
dimension	data dimension.
numData	number of data points.
numCenters	number of clusters.

Returns

K-means energy at the end of optimization.

The function initializes the centers by using the initialization algorithm set by vl_kmeans_set_initialization and refines them by the quantization algorithm set by vl_kmeans_set_algorithm. The process is repeated one or more times (see vl_kmeans_set_num_repetitions) and the resutl with smaller energy is retained.

vl_kmeans_delete()

void vl_kmeans_delete

(

VlKMeans *

self

)

Parameters

self	KMeans object instance.

The function deletes the KMeans object instance created by vl_kmeans_new.

vl_kmeans_init_centers_plus_plus()

void vl_kmeans_init_centers_plus_plus	(	VlKMeans *	self,
		void const *	data,
		vl_size	dimension,
		vl_size	numData,
		vl_size	numCenters
	)

Parameters

self	KMeans object.
data	data to sample from.
dimension	data dimension.
numData	nmber of data points.
numCenters	number of centers.

vl_kmeans_init_centers_with_rand_data()

void vl_kmeans_init_centers_with_rand_data	(	VlKMeans *	self,
		void const *	data,
		vl_size	dimension,
		vl_size	numData,
		vl_size	numCenters
	)

Parameters

self	KMeans object.
data	data to sample from.
dimension	data dimension.
numData	nmber of data points.
numCenters	number of centers.

The function inits the KMeans centers by randomly sampling the data data.

vl_kmeans_new()

VlKMeans* vl_kmeans_new	(	vl_type	dataType,
		VlVectorComparisonType	distance
	)

Parameters

dataType	type of data (VL_TYPE_FLOAT or VL_TYPE_DOUBLE)
distance	distance.

Returns

new KMeans object instance.

vl_kmeans_new_copy()

VlKMeans* vl_kmeans_new_copy

(

VlKMeans const *

kmeans

)

Parameters

kmeans

KMeans object to copy.

Returns

new copy.

vl_kmeans_quantize()

void vl_kmeans_quantize	(	VlKMeans *	self,
		vl_uint32 *	assignments,
		void *	distances,
		void const *	data,
		vl_size	numData
	)

Parameters

self	KMeans object.
assignments	data to closest center assignments (output).
distances	data to closest center distance (output).
data	data to quantize.
numData	number of data points to quantize.

vl_kmeans_quantize_ann()

void vl_kmeans_quantize_ann	(	VlKMeans *	self,
		vl_uint32 *	assignments,
		void *	distances,
		void const *	data,
		vl_size	numData,
		vl_bool	update
	)

Parameters

self	KMeans object.
assignments	data to centers assignments (output).
distances	data to closes center distance (output)
data	data to quantize.
numData	number of data points.
update	choose wether to update current assignments.

The function uses an ANN procedure to compute the approximate nearest neighbours of the input data point.

Setting update to VL_TRUE will cause the algorithm to update existing assignments. This means that each element of assignments and distances is updated ony if the ANN procedure can find a better assignment of the existing one.

vl_kmeans_refine_centers()

double vl_kmeans_refine_centers	(	VlKMeans *	self,
		void const *	data,
		vl_size	numData
	)

Parameters

self	KMeans object.
data	data to quantize.
numData	number of data points.

Returns

K-means energy at the end of optimization.

The function calls the underlying K-means quantization algorithm (VlKMeansAlgorithm) to quantize the specified data data. The function assumes that the cluster centers have already been assigned by using one of the seeding functions, or by setting them.

vl_kmeans_reset()

void vl_kmeans_reset

(

VlKMeans *

self

)

The function reset the state of the KMeans object. It deletes any stored centers, releasing the corresponding memory. This cancels the effect of seeding or setting the centers, but does not change the other configuration parameters.

vl_kmeans_set_centers()

void vl_kmeans_set_centers	(	VlKMeans *	self,
		void const *	centers,
		vl_size	dimension,
		vl_size	numCenters
	)

Parameters

self	KMeans object.
centers	centers to copy.
dimension	data dimension.
numCenters	number of centers.