public class SoftDTWModel extends Object implements AlignmentDerivativeAlgorithm
Constructor and Description |
---|
SoftDTWModel(AlignmentSpecification specificaton,
Sequence x,
Sequence y,
double[][] similarityMatrix) |
Modifier and Type | Method and Description |
---|---|
static double |
calc_weighted_comp_sum(Comparator[] comparators,
double[] weights,
int[] originalIndices,
Node nodeX,
Node nodeY)
Returns weighted sum k of local comparators: k = sum(w_i * k_i) with:
k_i: local comparators w_i: local weights
Note: Comparators, weights, originalIndices are supposed to originate
from the same AlignmentSpecification.
|
<X extends Value,Y> |
calculateParameterDerivative(DerivableComparator<X,Y> comp,
String keyword)
This returns the derivatives for all parameters of the given comparator
according to the implementation.
|
<X extends Value,Y> |
calculateRawParameterDerivative(DerivableComparator<X,Y> comp,
String keyword)
This returns the derivatives for all parameters of the given comparator
according to the implementation.
|
double[] |
calculateWeightDerivative()
Calculates the derivative for each keyword weight.
|
double |
getDistance()
Returns the distance between the two input sequences of this Derivative.
|
Sequence |
getLeft()
Returns the left sequence of this derivative.
|
Sequence |
getRight()
Returns the right sequence of this derivative.
|
double[][] |
getSimilarityMatrix()
This is the dynamic programming matrix calculated by a
DynamicTimeWarpingAlgorithm.
|
double |
getSimilarityScore()
Returns the softDTW score between the two input sequences.
|
AlignmentSpecification |
getSpecificaton()
Returns the AlignmentSpecification that was basis for this Alignment.
|
public SoftDTWModel(AlignmentSpecification specificaton, Sequence x, Sequence y, double[][] similarityMatrix)
public AlignmentSpecification getSpecificaton()
public double[][] getSimilarityMatrix()
public double getSimilarityScore()
public <X extends Value,Y> Y calculateParameterDerivative(DerivableComparator<X,Y> comp, String keyword)
calculateParameterDerivative
in interface AlignmentDerivativeAlgorithm
X
- the values the given comparator can compare.Y
- the result format for derivatives of the given comparator.comp
- the DerivableComparator that provides functionality to
calculate local derivatives.keyword
- the keyword for which the given comparator was used in
the Alignment.public <X extends Value,Y> double[] calculateRawParameterDerivative(DerivableComparator<X,Y> comp, String keyword)
calculateRawParameterDerivative
in interface AlignmentDerivativeAlgorithm
X
- the values the given comparator can compare.comp
- the DerivableComparator that provides functionality to
calculate local derivatives.keyword
- the keyword for which the given comparator was used in
the Alignment.public double[] calculateWeightDerivative()
calculateWeightDerivative
in interface AlignmentDerivativeAlgorithm
public static double calc_weighted_comp_sum(Comparator[] comparators, double[] weights, int[] originalIndices, Node nodeX, Node nodeY)
comparators
- local comparators k_i (from AlignmentSpecification)weights
- local weights w_i (from AlignmentSpecification)originalIndices
- indices (corresponding to comparators and weights)
to get correct node values (from AlignmentSpecification)nodeX
- first node to comparenodeY
- second node to comparepublic double getDistance()
getDistance
in interface AlignmentDerivativeAlgorithm
public Sequence getLeft()
getLeft
in interface AlignmentDerivativeAlgorithm
public Sequence getRight()
getRight
in interface AlignmentDerivativeAlgorithm
Copyright (C) 2013-2015 Benjamin Paaßen, Georg Zentgraf, AG Theoretical Computer Science, Centre of Excellence Cognitive Interaction Technology (CITEC), University of Bielefeld, licensed under the AGPL v. 3: http://openresearch.cit-ec.de/projects/tcs