Package elki.distance.timeseries

Class DerivativeDTWDistance

java.lang.Object

elki.distance.AbstractNumberVectorDistance

elki.distance.timeseries.AbstractEditDistance

elki.distance.timeseries.DTWDistance

elki.distance.timeseries.DerivativeDTWDistance

All Implemented Interfaces:

Distance<NumberVector>, NumberVectorDistance<NumberVector>, PrimitiveDistance<NumberVector>

@Title("Derivative dynamic time warping")
@Reference(authors="E. J. Keogh, M. J. Pazzani",
           title="Derivative dynamic time warping",
           booktitle="1st SIAM Int. Conf. on Data Mining (SDM-2001)",
           url="https://doi.org/10.1137/1.9781611972719.1",
           bibkey="DBLP:conf/sdm/KeoghP01")
public class DerivativeDTWDistance
extends DTWDistance

Derivative Dynamic Time Warping distance for numerical vectors.

Reference:

E. J. Keogh, M. J. Pazzani
Derivative dynamic time warping
1st SIAM Int. Conf. on Data Mining (SDM-2001)

Since:

0.7.0

Author:

Lara Hirschbeck, Daniel Kolb

Nested Class Summary

Nested classes/interfaces inherited from class elki.distance.timeseries.DTWDistance

DTWDistance.Par

Field Summary

Fields inherited from class elki.distance.timeseries.AbstractEditDistance

bandSize

Constructor Summary

Constructors

Constructor	Description
DerivativeDTWDistance()	Constructor.
DerivativeDTWDistance(double bandSize)	Constructor.

Method Summary

Modifier and Type	Method	Description
protected double	derivative(int i, NumberVector v)	Given a NumberVector and the position of an element, approximates the gradient of given element.
double	distance(NumberVector v1, NumberVector v2)	Computes the distance between two given vectors according to this distance function.
protected void	firstRow(double[] buf, int band, NumberVector v1, NumberVector v2, int dim2)	Fill the first row.

Methods inherited from class elki.distance.timeseries.DTWDistance

delta

Methods inherited from class elki.distance.timeseries.AbstractEditDistance

effectiveBandSize, equals, getInputTypeRestriction, hashCode

Methods inherited from class elki.distance.AbstractNumberVectorDistance

dimensionality, dimensionality, dimensionality, dimensionality, dimensionality, dimensionality, dimensionality, dimensionality

Methods inherited from class java.lang.Object

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

Methods inherited from interface elki.distance.Distance

isMetric, isSquared, isSymmetric

Methods inherited from interface elki.distance.PrimitiveDistance

instantiate

Constructor Detail

DerivativeDTWDistance

public DerivativeDTWDistance()

Constructor.

DerivativeDTWDistance

public DerivativeDTWDistance(double bandSize)

Constructor.

Parameters:

bandSize - Band size

Method Detail

distance

public double distance(NumberVector v1,
                       NumberVector v2)

Description copied from interface: NumberVectorDistance

Computes the distance between two given vectors according to this distance function.

Specified by:

distance in interface NumberVectorDistance<NumberVector>

Specified by:

distance in interface PrimitiveDistance<NumberVector>

Overrides:

distance in class DTWDistance

Parameters:

v1 - first vector

v2 - second vector

Returns:

the distance between two given vectors according to this distance function

firstRow

protected void firstRow(double[] buf,
                        int band,
                        NumberVector v1,
                        NumberVector v2,
                        int dim2)

Description copied from class: DTWDistance

Fill the first row.

Overrides:

firstRow in class DTWDistance

Parameters:

buf - Buffer

band - Bandwidth

v1 - First vector

v2 - Second vector

dim2 - Dimensionality of second

derivative

protected double derivative(int i,
                            NumberVector v)

Given a NumberVector and the position of an element, approximates the gradient of given element.

Returns:

Derivative as double