org.openimaj.ml.clustering.kmeans

Class HierarchicalLongKMeans

java.lang.Object

org.openimaj.ml.clustering.kmeans.HierarchicalLongKMeans

All Implemented Interfaces:

Clusterer<long[][]>, SpatialClusterer<HierarchicalLongKMeansResult,long[]>

@Reference(type=Inproceedings,
           author={"David. Nist\'er","Henrik. Stew\'enius"},
           title="Scalable Recognition with a Vocabulary Tree",
           year="2006",
           booktitle="CVPR",
           pages={"2161","","2168"},
           customData={"Date-Added","2010-11-12 09:33:18 +0000","Date-Modified","2010-11-22 15:11:22 +0000"})
public class HierarchicalLongKMeans
extends Object
implements SpatialClusterer<HierarchicalLongKMeansResult,long[]>

Hierarchical Long K-Means clustering (HierarchicalLongKMeans) is a simple hierarchical version of LongKMeans. The algorithm recursively applies

Author:

Sina Samangooei (ss@ecs.soton.ac.uk), Jonathon Hare (jsh2@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
HierarchicalLongKMeans(int M, int K, int depth) Construct a new HierarchicalLongKMeans with the given parameters.
HierarchicalLongKMeans(KMeansConfiguration<LongNearestNeighbours,long[]> config, int M, int K, int depth) Construct a new HierarchicalLongKMeans with the given parameters.

Method Summary

Modifier and Type	Method and Description
HierarchicalLongKMeansResult	cluster(DataSource<long[]> data) Perform clustering with data from a data source.
HierarchicalLongKMeansResult	cluster(long[][] data) Perform clustering on the given data.
int[][]	performClustering(long[][] data)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

HierarchicalLongKMeans

public HierarchicalLongKMeans(KMeansConfiguration<LongNearestNeighbours,long[]> config,
                              int M,
                              int K,
                              int depth)

Construct a new HierarchicalLongKMeans with the given parameters.

Parameters:

config - configuration for the underlying kmeans clustering.

M - Data dimensionality.

K - Number of clusters per node.

depth - Tree depth.

HierarchicalLongKMeans

public HierarchicalLongKMeans(int M,
                              int K,
                              int depth)

Construct a new HierarchicalLongKMeans with the given parameters. Uses the default parameters of the KMeansConfiguration.

Parameters:

M - Data dimensionality.

K - Number of clusters per node.

depth - Tree depth.

Method Detail

cluster

public HierarchicalLongKMeansResult cluster(long[][] data)

Description copied from interface: SpatialClusterer

Perform clustering on the given data.

Specified by:

cluster in interface SpatialClusterer<HierarchicalLongKMeansResult,long[]>

Parameters:

data - the data.

Returns:

the generated clusters.

performClustering

public int[][] performClustering(long[][] data)

Specified by:

performClustering in interface Clusterer<long[][]>

Returns:

Given data items, cluster them by index

cluster

public HierarchicalLongKMeansResult cluster(DataSource<long[]> data)

Description copied from interface: SpatialClusterer

Perform clustering with data from a data source. The DataSource could potentially be backed by disk rather in memory.

Specified by:

cluster in interface SpatialClusterer<HierarchicalLongKMeansResult,long[]>

Parameters:

data - the data.

Returns:

the generated clusters.