File statistics.hpp¶

class IStatistics¶

#include <statistics.hpp>

Defines an interface for all classes that provide access to statistics about the quality of predictions for training examples, which serve as the basis for learning a new rule or refining an existing one.

Public Functions

inline virtual ~IStatistics()¶

virtual uint32 getNumStatistics() const = 0¶

Returns the number of available statistics.

Returns:: The number of statistics

virtual uint32 getNumOutputs() const = 0¶

Returns the number of available outputs.

Returns:: The number of outputs

virtual void applyPrediction(uint32 statisticIndex, const CompletePrediction &prediction) = 0¶

Updates a specific statistic based on the prediction of a rule that predicts for all available outputs.

This function must be called for each statistic that is covered by the new rule before learning the next rule.

Parameters:

statisticIndex – The index of the statistic to be updated
prediction – A reference to an object of type CompletePrediction that stores the scores that are predicted by the rule

virtual void applyPrediction(uint32 statisticIndex, const PartialPrediction &prediction) = 0¶

Updates a specific statistic based on the prediction of a rule that predicts for a subset of the available outputs.

This function must be called for each statistic that is covered by the new rule before learning the next rule.

Parameters:

statisticIndex – The index of the statistic to be updated
prediction – A reference to an object of type PartialPrediction that stores the scores that are predicted by the rule

virtual void revertPrediction(uint32 statisticIndex, const CompletePrediction &prediction) = 0¶

Reverts a specific statistic that has previously been updated via the function applyPrediction based on the prediction of a rule that predicts for all available outputs.

Parameters:

statisticIndex – The index of the statistic to be reverted
prediction – A reference to an object of type CompletePrediction that stores the scores that are predicted by the rule

virtual void revertPrediction(uint32 statisticIndex, const PartialPrediction &prediction) = 0¶

Reverts a specific statistic that has previously been updated via the function applyPrediction based on the prediction of a rule that predicts for a subset of the available outputs.

Parameters:

statisticIndex – The index of the statistic to be reverted
prediction – A reference to an object of type PartialPrediction that stores the scores that are predicted by the rule

virtual float64 evaluatePrediction(uint32 statisticIndex) const = 0¶

Calculates and returns a numerical score that assesses the quality of the current predictions for a specific statistic.

Parameters:: statisticIndex – The index of the statistic for which the predictions should be evaluated
Returns:: The numerical score that has been calculated

virtual std::unique_ptr<IStatisticsSubset> createSubset(const CompleteIndexVector &outputIndices, const EqualWeightVector &weights) const = 0¶

Creates and returns a new object of type IStatisticsSubset that includes only those outputs, whose indices are provided by a specific CompleteIndexVector.

Parameters:

outputIndices – A reference to an object of type CompleteIndexVector that provides access to the indices of the outputs that should be included in the subset
weights – A reference to an object of type EqualWeightVector that provides access to the weights of individual training examples

Returns: