File rule_evaluation_example_wise_partial_fixed.hpp

namespace boosting
class ExampleWiseFixedPartialRuleEvaluationFactory : public boosting::IExampleWiseRuleEvaluationFactory
#include <rule_evaluation_example_wise_partial_fixed.hpp>

Allows to create instances of the class IExampleWiseRuleEvaluationFactory that allow to calculate the predictions of partial rules, which predict for a predefined number of labels.

Public Functions

ExampleWiseFixedPartialRuleEvaluationFactory(float32 labelRatio, uint32 minLabels, uint32 maxLabels, float64 l1RegularizationWeight, float64 l2RegularizationWeight, const Blas &blas, const Lapack &lapack)
Parameters:

  • labelRatio – A percentage that specifies for how many labels the rule heads should predict, e.g., if 100 labels are available, a percentage of 0.5 means that the rule heads predict for a subset of ceil(0.5 * 100) = 50 labels. Must be in (0, 1)

  • minLabels – The minimum number of labels for which the rule heads should predict. Must be at least 2

  • maxLabels – The maximum number of labels for which the rule heads should predict. Must be at least minLabels or 0, if the maximum number of labels should not be restricted

  • l1RegularizationWeight – The weight of the L1 regularization that is applied for calculating the scores to be predicted by rules

  • l2RegularizationWeight – The weight of the L2 regularization that is applied for calculating the scores to be predicted by rules

  • blas – A reference to an object of type Blas that allows to execute BLAS routines

  • lapack – An reference to an object of type Lapack that allows to execute BLAS routines

virtual std::unique_ptr<IRuleEvaluation<DenseExampleWiseStatisticVector>> create(const DenseExampleWiseStatisticVector &statisticVector, const CompleteIndexVector &indexVector) const override

Creates and returns a new object of type IRuleEvaluation that allows to calculate the predictions of rules that predict for all available labels, based on the gradients and Hessians that are stored by a DenseExampleWiseStatisticVector.

Parameters:

  • statisticVector – A reference to an object of type DenseExampleWiseStatisticVector. This vector is only used to identify the function that is able to deal with this particular type of vector via function overloading

  • indexVector – A reference to an object of type CompleteIndexVector that provides access to the indices of the labels for which the rules may predict

Returns:

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

virtual std::unique_ptr<IRuleEvaluation<DenseExampleWiseStatisticVector>> create(const DenseExampleWiseStatisticVector &statisticVector, const PartialIndexVector &indexVector) const override

Creates and returns a new object of type IRuleEvaluation that allows to calculate the predictions of rules that predict for a subset of the available labels, based on the gradients and Hessians that are stored by a DenseExampleWiseStatisticVector.

Parameters:

  • statisticVector – A reference to an object of type DenseExampleWiseStatisticVector. This vector is only used to identify the function that is able to deal with this particular type of vector via function overloading

  • indexVector – A reference to an object of type PartialIndexVector that provides access to the indices of the labels for which the rules may predict

Returns:

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

Private Members

const float32 labelRatio_
const uint32 minLabels_
const uint32 maxLabels_
const float64 l1RegularizationWeight_
const float64 l2RegularizationWeight_
const Blas &blas_
const Lapack &lapack_