File probability_function_marginal.hpp

namespace boosting
class IMarginalProbabilityFunction
#include <probability_function_marginal.hpp>

Defines an interface for all classes that allow to transform regression scores that are predicted for individual labels into marginal probabilities.

Public Functions

inline virtual ~IMarginalProbabilityFunction()
virtual float64 transformScoreIntoMarginalProbability(uint32 labelIndex, float64 score) const = 0

Transforms the regression score that is predicted for a specific label into a probability.

Parameters:

  • labelIndex – The index of the label, the regression score is predicted for

  • score – The regression score that is predicted

Returns:

The probability into which the given score was transformed

class IMarginalProbabilityFunctionFactory
#include <probability_function_marginal.hpp>

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

Subclassed by boosting::LogisticFunctionFactory

Public Functions

inline virtual ~IMarginalProbabilityFunctionFactory()
virtual std::unique_ptr<IMarginalProbabilityFunction> create(const IMarginalProbabilityCalibrationModel &marginalProbabilityCalibrationModel) const = 0

Creates and returns a new object of the type IMarginalProbabilityFunction.

Parameters:

marginalProbabilityCalibrationModel – A reference to an object of type IMarginalProbabilityCalibrationModel that should be used for the calibration of marginal probabilities

Returns:

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