Package elki.math.statistics.intrinsicdimensionality

Interface DistanceBasedIntrinsicDimensionalityEstimator

All Superinterfaces:

IntrinsicDimensionalityEstimator<java.lang.Object>

All Known Implementing Classes:

ABIDEstimator, AggregatedHillEstimator, EnsembleEstimator, GEDEstimator, HillEstimator, LMomentsEstimator, MOMEstimator, PWM2Estimator, PWMEstimator, RABIDEstimator, RVEstimator, ZipfEstimator

public interface DistanceBasedIntrinsicDimensionalityEstimator
extends IntrinsicDimensionalityEstimator<java.lang.Object>

Distance-based ID estimator.

This is the common case, but we have some estimators that require either the ability to query for neighbors of additional points, or to access the coordinate data.

Since:

0.7.0

Author:

Erich Schubert

Method Summary

Modifier and Type	Method	Description
static <A> int	countLeadingZeros(A data, NumberArrayAdapter<?,? super A> adapter, int end)
default double	estimate(double[] distances)	Estimate from a distance list.
default double	estimate(double[] distances, int size)	Estimate from a distance list.
default <A> double	estimate(A data, NumberArrayAdapter<?,? super A> adapter)	Estimate from a distance list.
<A> double	estimate(A data, NumberArrayAdapter<?,? super A> adapter, int size)	Estimate from a distance list.
default double	estimate(KNNSearcher<DBIDRef> knnq, DistanceQuery<? extends java.lang.Object> distq, DBIDRef cur, int k)	Estimate from a Reference Point, a KNNSearcher and the neighborhood size k.
default double	estimate(RangeSearcher<DBIDRef> rnq, DistanceQuery<? extends java.lang.Object> distq, DBIDRef cur, double range)	Estimate from a distance list.

Method Detail

estimate

<A> double estimate(A data,
                    NumberArrayAdapter<?,? super A> adapter,
                    int size)

Estimate from a distance list.

Type Parameters:

A - array type

Parameters:

data - Data

adapter - Array adapter

size - Length

Returns:

Estimated intrinsic dimensionality

estimate

default double estimate(double[] distances)

Estimate from a distance list.

Parameters:

distances - Distances

Returns:

Estimated intrinsic dimensionality

estimate

default double estimate(double[] distances,
                        int size)

Estimate from a distance list.

Parameters:

distances - Distances

size - Valid size

Returns:

Estimated intrinsic dimensionality

estimate

default <A> double estimate(A data,
                            NumberArrayAdapter<?,? super A> adapter)

Estimate from a distance list.

Type Parameters:

A - array type

Parameters:

data - Data

adapter - Array adapter

Returns:

Estimated intrinsic dimensionality

estimate

default double estimate(KNNSearcher<DBIDRef> knnq,
                        DistanceQuery<? extends java.lang.Object> 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<java.lang.Object>

Parameters:

knnq - KNNSearcher

distq - Distance query for additional distances

cur - reference point

k - neighborhood size

Returns:

Estimated intrinsic dimensionality

estimate

default double estimate(RangeSearcher<DBIDRef> rnq,
                        DistanceQuery<? extends java.lang.Object> distq,
                        DBIDRef cur,
                        double range)

Description copied from interface: IntrinsicDimensionalityEstimator

Estimate from a distance list.

Specified by:

estimate in interface IntrinsicDimensionalityEstimator<java.lang.Object>

Parameters:

rnq - RangeSearcher

distq - Distance query for additional distances

cur - reference point

range - neighborhood radius

Returns:

Estimated intrinsic dimensionality

countLeadingZeros

static <A> int countLeadingZeros(A data,
                                 NumberArrayAdapter<?,? super A> adapter,
                                 int end)

Parameters:

data - Data array

adapter - Adapter class

end - Length

Returns:

Number of leading zero distances.