Package elki.math.statistics.intrinsicdimensionality

Class LPCAEstimator

java.lang.Object

elki.math.statistics.intrinsicdimensionality.LPCAEstimator

All Implemented Interfaces:

IntrinsicDimensionalityEstimator<NumberVector>

public class LPCAEstimator
extends java.lang.Object
implements IntrinsicDimensionalityEstimator<NumberVector>

Local PCA based ID estimator.

Since:

0.8.0

Author:

Erik Thordsen

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	LPCAEstimator.Par	Parameterization class.

Field Summary

Fields

Modifier and Type	Field	Description
protected EigenPairFilter	eigenFilter	Eigenvalue filter
protected PCARunner	pcaRunner	Class to perform PCA

Constructor Summary

Constructors

Constructor	Description
LPCAEstimator(EigenPairFilter eigenFilter)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected double	estimate(DBIDs ids, Relation<? extends NumberVector> relation)	Returns an ID estimate based on the specified filter for the given point DBID set and relation.
double	estimate(KNNSearcher<DBIDRef> knnq, DistanceQuery<? extends NumberVector> distq, DBIDRef cur, int k)	Estimate from a Reference Point, a KNNSearcher and the neighborhood size k.
double	estimate(RangeSearcher<DBIDRef> rnq, DistanceQuery<? extends NumberVector> distq, DBIDRef cur, double range)	Estimate from a distance list.

Methods inherited from class java.lang.Object

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

Field Detail

pcaRunner

protected PCARunner pcaRunner

Class to perform PCA

eigenFilter

protected EigenPairFilter eigenFilter

Eigenvalue filter

Constructor Detail

LPCAEstimator

public LPCAEstimator(EigenPairFilter eigenFilter)

Constructor.

Parameters:

eigenFilter - Filter to choose the number of eigenvalues to keep

Method Detail

estimate

public double estimate(KNNSearcher<DBIDRef> knnq,
                       DistanceQuery<? extends NumberVector> distq,
                       DBIDRef cur,
                       int k)

Description copied from interface: IntrinsicDimensionalityEstimator

Estimate from a Reference Point, a KNNSearcher and the neighborhood size k.

Specified by:

estimate in interface IntrinsicDimensionalityEstimator<NumberVector>

Parameters:

knnq - KNNSearcher

distq - Distance query for additional distances

cur - reference point

k - neighborhood size

Returns:

Estimated intrinsic dimensionality

estimate

public double estimate(RangeSearcher<DBIDRef> rnq,
                       DistanceQuery<? extends NumberVector> distq,
                       DBIDRef cur,
                       double range)

Description copied from interface: IntrinsicDimensionalityEstimator

Estimate from a distance list.

Specified by:

estimate in interface IntrinsicDimensionalityEstimator<NumberVector>

Parameters:

rnq - RangeSearcher

distq - Distance query for additional distances

cur - reference point

range - neighborhood radius

Returns:

Estimated intrinsic dimensionality

estimate

protected double estimate(DBIDs ids,
                          Relation<? extends NumberVector> relation)

Returns an ID estimate based on the specified filter for the given point DBID set and relation.

Parameters:

ids - neighbor objects

relation - data vector relation

Returns:

estimated intrinsic dimensionality