File measure_distance.hpp

class IDistanceMeasure
#include <measure_distance.hpp>

Defines an interface for all measures that may be used to compare predictions for individual examples to the corresponding ground truth labels in order to obtain a distance.

Public Functions

inline virtual ~IDistanceMeasure()
virtual float64 measureDistance(uint32 labelVectorIndex, const LabelVector &labelVector, View<float64>::const_iterator scoresBegin, View<float64>::const_iterator scoresEnd) const = 0

Calculates and returns the distance between the predicted scores for a single example and a label vector.

Parameters:

  • labelVectorIndex – The index of the label vector, the scores should be compared to

  • labelVector – A reference to an object of type LabelVector, the scores should be compared to

  • scoresBegin – An iterator to the beginning of the predicted scores

  • scoresEnd – An iterator to the end of the predicted scores

Returns:

The distance that has been calculated

inline virtual const LabelVector &getClosestLabelVector(const LabelVectorSet &labelVectorSet, View<float64>::const_iterator scoresBegin, View<float64>::const_iterator scoresEnd) const

Searches among the label vectors contained in a LabelVectorSet and returns the one that is closest to the scores that are predicted for an example.

Parameters:

  • labelVectorSet – A reference to an object of type LabelVectorSet that contains the label vectors

  • scoresBegin – An iterator to the beginning of the predicted scores

  • scoresEnd – An iterator to the end of the predicted scores

Returns:

A reference to an object of type LabelVector that has been found

class IDistanceMeasureFactory
#include <measure_distance.hpp>

Defines an interface for all factories that allow to create instances of the type IDistanceMeasure.

Public Functions

inline virtual ~IDistanceMeasureFactory()
virtual std::unique_ptr<IDistanceMeasure> createDistanceMeasure(const IMarginalProbabilityCalibrationModel &marginalProbabilityCalibrationModel, const IJointProbabilityCalibrationModel &jointProbabilityCalibrationModel) const = 0

Creates and returns a new object of type IDistanceMeasure.

Parameters:
Returns:

An unique pointer to an object of type IDistanceMeasure that has been created