Package elki.algorithm.statistics

Class DistanceStatisticsWithClasses<O>

java.lang.Object

elki.algorithm.statistics.DistanceStatisticsWithClasses<O>

Type Parameters:

O - Object type

All Implemented Interfaces:

Algorithm

@Title("Distance Histogram")
@Description("Computes a histogram over the distances occurring in the data set.")
public class DistanceStatisticsWithClasses<O>
extends java.lang.Object
implements Algorithm

Algorithm to gather statistics over the distance distribution in the data set.

Since:

0.2

Author:

Erich Schubert

Nested Class Summary

Nested classes/interfaces inherited from interface elki.Algorithm

Algorithm.Utils

Field Summary

Fields

Modifier and Type	Field	Description
protected Distance<? super O>	distance	Distance function used.
protected boolean	exact	Compute exactly (slower).
private static Logging	LOG	The logger for this class.
protected int	numbin	Number of bins to use in sampling.
protected boolean	sampling	Sampling flag.

Constructor Summary

Constructors

Constructor	Description
DistanceStatisticsWithClasses(Distance<? super O> distance, int numbins, boolean exact, boolean sampling)	Constructor.

Method Summary

Modifier and Type	Method	Description
private DoubleMinMax	exactMinMax(Relation<O> relation, DistanceQuery<O> distance)	Compute the exact maximum and minimum.
TypeInformation[]	getInputTypeRestriction()	Get the input type restriction used for negotiating the data query.
HistogramResult	run(Database database, Relation<O> relation)
private DoubleMinMax	sampleMinMax(Relation<O> relation, DistanceQuery<O> distance)	Estimate minimum and maximum via sampling.
private static void	shrinkHeap(java.util.TreeSet<DoubleDBIDPair> hotset, int k)	Shrink the heap of "hot" (extreme) items.

Methods inherited from class java.lang.Object

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

Methods inherited from interface elki.Algorithm

autorun

Field Detail

LOG

private static final Logging LOG

The logger for this class.

distance

protected Distance<? super O> distance

Distance function used.

numbin

protected int numbin

Number of bins to use in sampling.

sampling

protected boolean sampling

Sampling flag.

exact

protected boolean exact

Compute exactly (slower).

Constructor Detail

DistanceStatisticsWithClasses

public DistanceStatisticsWithClasses(Distance<? super O> distance,
                                     int numbins,
                                     boolean exact,
                                     boolean sampling)

Constructor.

Parameters:

distance - Distance function to use

numbins - Number of bins

exact - Exactness flag

sampling - Sampling flag

Method Detail

getInputTypeRestriction

public TypeInformation[] getInputTypeRestriction()

Description copied from interface: Algorithm

Get the input type restriction used for negotiating the data query.

Specified by:

getInputTypeRestriction in interface Algorithm

Returns:

Type restriction

run

public HistogramResult run(Database database,
                           Relation<O> relation)

sampleMinMax

private DoubleMinMax sampleMinMax(Relation<O> relation,
                                  DistanceQuery<O> distance)

Estimate minimum and maximum via sampling.

Parameters:

relation - Relation to process

distance - Distance function to use

Returns:

Minimum and maximum

exactMinMax

private DoubleMinMax exactMinMax(Relation<O> relation,
                                 DistanceQuery<O> distance)

Compute the exact maximum and minimum.

Parameters:

relation - Relation to process

distance - Distance function

Returns:

Exact maximum and minimum

shrinkHeap

private static void shrinkHeap(java.util.TreeSet<DoubleDBIDPair> hotset,
                               int k)

Shrink the heap of "hot" (extreme) items.

Parameters:

hotset - Set of hot items

k - target size