File partition_sampling_auto.hpp¶

namespace boosting

class AutomaticPartitionSamplingConfig : public IPartitionSamplingConfig ¶

#include <partition_sampling_auto.hpp>

Allows to configure a method that automatically decides for a method that partitions the available training examples into a training set and a holdout set, depending on whether a holdout set is needed and depending on the loss function.

Public Functions

AutomaticPartitionSamplingConfig(const std::unique_ptr<IGlobalPruningConfig> &globalPruningConfigPtr, const std::unique_ptr<IMarginalProbabilityCalibratorConfig> &marginalProbabilityCalibratorConfigPtr, const std::unique_ptr<IJointProbabilityCalibratorConfig> &jointProbabilityCalibratorConfigPtr)¶

Parameters:

globalPruningConfigPtr – A reference to an unique pointer that stores the configuration of the method that is used for pruning entire rules
marginalProbabilityCalibratorConfigPtr – A reference to an unique pointer that stores the configuration of the calibrator that is used to fit a model for the calibration of marginal probabilities
jointProbabilityCalibratorConfigPtr – A reference to an unique pointer that stores the configuration of the calibrator that is used to fit a model for the calibration of joint probabilities

std::unique_ptr<IPartitionSamplingFactory> createPartitionSamplingFactory() const override¶