Package elki.index.distancematrix

Class PrecomputedDistanceMatrix<O>

java.lang.Object

elki.index.distancematrix.PrecomputedDistanceMatrix<O>

Type Parameters:

O - Object type

All Implemented Interfaces:

elki.index.DistanceIndex<O>, elki.index.DistancePriorityIndex<O>, elki.index.Index, elki.index.KNNIndex<O>, elki.index.RangeIndex<O>

public class PrecomputedDistanceMatrix<O>
extends java.lang.Object
implements elki.index.DistanceIndex<O>, elki.index.RangeIndex<O>, elki.index.KNNIndex<O>, elki.index.DistancePriorityIndex<O>

Distance matrix, for precomputing similarity for a small data set.

This class uses a linear memory layout (not a ragged array), and assumes symmetry as well as strictness. This way, it only stores the upper triangle matrix with double precision. It has to store (n-1) * (n-2) distance values in memory, requiring 8 * (n-1) * (n-2) bytes. Since Java has a size limit of arrays of 31 bits (signed integer), we can store at most \(2^{16}\) objects (precisely, 65536 objects) in a single array, which needs about 16 GB of RAM.

Since:

0.7.0

Author:

Erich Schubert

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	PrecomputedDistanceMatrix.Factory<O>	Factory for the index.
class	PrecomputedDistanceMatrix.PrecomputedDistancePrioritySearcher	Range query using the distance matrix.
class	PrecomputedDistanceMatrix.PrecomputedDistanceQuery	Distance query using the precomputed matrix.
class	PrecomputedDistanceMatrix.PrecomputedKNNQuery	kNN query using the distance matrix.
class	PrecomputedDistanceMatrix.PrecomputedRangeQuery	Range query using the distance matrix.

Field Summary

Fields

Modifier and Type	Field	Description
protected elki.distance.Distance<? super O>	distance	Nested distance function.
private elki.database.ids.DBIDRange	ids	DBID range.
private static elki.logging.Logging	LOG	Class logger.
private double[]	matrix	Distance matrix.
protected java.lang.ref.WeakReference<elki.database.relation.Relation<O>>	refrelation	Data relation.

Constructor Summary

Constructors

Constructor	Description
PrecomputedDistanceMatrix(elki.database.relation.Relation<O> relation, elki.database.ids.DBIDRange range, elki.distance.Distance<? super O> distance)	Constructor.

Method Summary

Modifier and Type	Method	Description
elki.database.query.distance.DistanceQuery<O>	getDistanceQuery(elki.distance.Distance<? super O> distanceFunction)
private int	getOffset(int x, int y)	Array offset computation.
void	initialize()
elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef>	kNNByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery, int maxk, int flags)
elki.database.query.knn.KNNSearcher<O>	kNNByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery, int maxk, int flags)
void	logStatistics()
elki.database.query.PrioritySearcher<elki.database.ids.DBIDRef>	priorityByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery, double maxrange, int flags)
elki.database.query.PrioritySearcher<O>	priorityByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery, double maxrange, int flags)
elki.database.query.range.RangeSearcher<elki.database.ids.DBIDRef>	rangeByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery, double maxrange, int flags)
elki.database.query.range.RangeSearcher<O>	rangeByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery, double maxrange, int flags)
protected static int	triangleSize(int x)	Compute the size of a complete x by x triangle (minus diagonal)

Methods inherited from class java.lang.Object

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

Field Detail

LOG

private static final elki.logging.Logging LOG

Class logger.

refrelation

protected final java.lang.ref.WeakReference<elki.database.relation.Relation<O>> refrelation

Data relation.

distance

protected final elki.distance.Distance<? super O> distance

Nested distance function.

matrix

private double[] matrix

Distance matrix.

ids

private elki.database.ids.DBIDRange ids

DBID range.

Constructor Detail

PrecomputedDistanceMatrix

public PrecomputedDistanceMatrix(elki.database.relation.Relation<O> relation,
                                 elki.database.ids.DBIDRange range,
                                 elki.distance.Distance<? super O> distance)

Constructor.

Parameters:

relation - Data relation

range - DBID range

distance - Distance function

Method Detail

initialize

public void initialize()

Specified by:

initialize in interface elki.index.Index

triangleSize

protected static int triangleSize(int x)

Compute the size of a complete x by x triangle (minus diagonal)

Parameters:

x - Offset

Returns:

Size of complete triangle

getOffset

private int getOffset(int x,
                      int y)

Array offset computation.

Parameters:

x - X parameter

y - Y parameter

Returns:

Array offset

logStatistics

public void logStatistics()

Specified by:

logStatistics in interface elki.index.Index

getDistanceQuery

public elki.database.query.distance.DistanceQuery<O> getDistanceQuery(elki.distance.Distance<? super O> distanceFunction)

Specified by:

getDistanceQuery in interface elki.index.DistanceIndex<O>

kNNByObject

public elki.database.query.knn.KNNSearcher<O> kNNByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                          int maxk,
                                                          int flags)

Specified by:

kNNByObject in interface elki.index.DistancePriorityIndex<O>

Specified by:

kNNByObject in interface elki.index.KNNIndex<O>

kNNByDBID

public elki.database.query.knn.KNNSearcher<elki.database.ids.DBIDRef> kNNByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                                                int maxk,
                                                                                int flags)

Specified by:

kNNByDBID in interface elki.index.KNNIndex<O>

rangeByObject

public elki.database.query.range.RangeSearcher<O> rangeByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                                double maxrange,
                                                                int flags)

Specified by:

rangeByObject in interface elki.index.DistancePriorityIndex<O>

Specified by:

rangeByObject in interface elki.index.RangeIndex<O>

rangeByDBID

public elki.database.query.range.RangeSearcher<elki.database.ids.DBIDRef> rangeByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                                                      double maxrange,
                                                                                      int flags)

Specified by:

rangeByDBID in interface elki.index.RangeIndex<O>

priorityByObject

public elki.database.query.PrioritySearcher<O> priorityByObject(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                                double maxrange,
                                                                int flags)

Specified by:

priorityByObject in interface elki.index.DistancePriorityIndex<O>

priorityByDBID

public elki.database.query.PrioritySearcher<elki.database.ids.DBIDRef> priorityByDBID(elki.database.query.distance.DistanceQuery<O> distanceQuery,
                                                                                      double maxrange,
                                                                                      int flags)

Specified by:

priorityByDBID in interface elki.index.DistancePriorityIndex<O>