File probability_calibration.hpp

template<typename ProbabilityCalibrationModel>
class IProbabilityCalibrator
#include <probability_calibration.hpp>

Defines an interface for all classes that implement a method for fitting models for the calibration of probabilities.

Template Parameters:

ProbabilityCalibrationModel – The type of the calibration model that is fitted by the calibrator

Public Functions

inline virtual ~IProbabilityCalibrator()
virtual std::unique_ptr<ProbabilityCalibrationModel> fitProbabilityCalibrationModel(const SinglePartition &partition, const CContiguousView<const uint8> &labelMatrix, const IStatistics &statistics) const = 0

Fits and returns a model for the calibration of probabilities.

Parameters:

  • partition – A reference to an object of type SinglePartition that provides access to the indices of the training examples that are included in the training set

  • labelMatrix – A reference to an object of type CContiguousView that provides row-wise access to the labels of the training examples

  • statistics – A reference to an object of type IStatistics that provides access to statistics about the labels of the training examples

Returns:

An unique pointer to an object of template type ProbabilityCalibrationModel that has been fit

virtual std::unique_ptr<ProbabilityCalibrationModel> fitProbabilityCalibrationModel(const SinglePartition &partition, const BinaryCsrView &labelMatrix, const IStatistics &statistics) const = 0

Fits and returns a model for the calibration of probabilities.

Parameters:

  • partition – A reference to an object of type SinglePartition that provides access to the indices of the training examples that are included in the training set

  • labelMatrix – A reference to an object of type BinaryCsrView that provides row-wise access to the labels of the training examples

  • statistics – A reference to an object of type IStatistics that provides access to statistics about the labels of the training examples

Returns:

An unique pointer to an object of template type ProbabilityCalibrationModel that has been fit

virtual std::unique_ptr<ProbabilityCalibrationModel> fitProbabilityCalibrationModel(BiPartition &partition, const CContiguousView<const uint8> &labelMatrix, const IStatistics &statistics) const = 0

Fits and returns a model for the calibration of probabilities.

Parameters:

  • partition – A reference to an object of type BiPartition that provides access to the indices of the training examples that are included in the training set and the holdout set, respectively

  • labelMatrix – A reference to an object of type CContiguousView that provides row-wise access to the labels of the training examples

  • statistics – A reference to an object of type IStatistics that provides access to statistics about the labels of the training examples

Returns:

An unique pointer to an object of template type ProbabilityCalibrationModel that has been fit

virtual std::unique_ptr<ProbabilityCalibrationModel> fitProbabilityCalibrationModel(BiPartition &partition, const BinaryCsrView &labelMatrix, const IStatistics &statistics) const = 0

Fits and returns a model for the calibration of probabilities.

Parameters:

  • partition – A reference to an object of type BiPartition that provides access to the indices of the training examples that are included in the training set and the holdout set, respectively

  • labelMatrix – A reference to an object of type BinaryCsrView that provides row-wise access to the labels of the training examples

  • statistics – A reference to an object of type IStatistics that provides access to statistics about the labels of the training examples

Returns:

An unique pointer to an object of template type ProbabilityCalibrationModel that has been fit