File statistics_subset.hpp¶

Functions

template<typename StatisticVector, typename StatisticView> static inline void addStatisticsToVector(StatisticVector &statisticVector, const EqualWeightVector &weights, const StatisticView &statisticView, uint32 row)¶

Adds the statistics at a specific row of a view to a given vector. The statistics are weighted according to given weights.

Template Parameters:

StatisticVector – The type of the vector to be modified
StatisticView – The type of the view that provides access to the statistics

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be modified
weights – A reference to an object of type EqualWeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics
row – The index of the row in the view to be added to the vector

template<typename StatisticVector, typename WeightVector, typename StatisticView> static inline void addStatisticsToVector(StatisticVector &statisticVector, const WeightVector &weights, const StatisticView &statisticView, uint32 row)¶

Adds the statistics at a specific row of a view to a given vector. The statistics are weighted according to given weights.

Template Parameters:

StatisticVector – The type of the vector to be modified
WeightVector – The type of the vector that provides access to the weights of statistics
StatisticView – The type of the view that provides access to the statistics

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be modified
weights – A reference to an object of template type WeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics
row – The index of the row in the view to be added to the vector

template<typename StatisticVector, typename StatisticView> static inline void removeStatisticsFromVector(StatisticVector &statisticVector, const EqualWeightVector &weights, const StatisticView &statisticView, uint32 row)¶

Removes the statistics at a specific row of a view from a given vector. The statistics are weighted according to given weights.

Template Parameters:

StatisticVector – The type of the vector to be modified
StatisticView – The type of the view that provides access to the statistics

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be modified
weights – A reference to an object of type EqualWeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics
row – The index of the row in the view to be removed from the vector

template<typename StatisticVector, typename WeightVector, typename StatisticView> static inline void removeStatisticsFromVector(StatisticVector &statisticVector, const WeightVector &weights, const StatisticView &statisticView, uint32 row)¶

Removes the statistics at a specific row of a view from a given vector. The statistics are weighted according to given weights.

Template Parameters:

StatisticVector – The type of the vector to be modified
WeightVector – The type of the vector that provides access to the weights of statistics
StatisticView – The type of the view that provides access to the statistics

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be modified
weights – A reference to an object of template type WeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics
row – The index of the row in the view to be removed from the vector

template<typename StatisticVector, typename WeightVector, typename StatisticView, typename IndexIterator> static inline void removeStatisticsFromVector(StatisticVector &statisticVector, const WeightVector &weights, const StatisticView &statisticView, IndexIterator indicesBegin, IndexIterator indicesEnd)¶

Removes the statistics at specific rows of a view from a given vector. The statistics are weighted according to given weights.

Template Parameters:

StatisticVector – The type of the vector to be modified
WeightVector – The type of the vector that provides access to the weights of statistics
StatisticView – The type of the view that provides access to the statistics
IndexIterator – The type of the iterator that allows to access the indices of the rows

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be modified
weights – A reference to an object of template type WeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics
indicesBegin – An iterator to the binning of the indices
indicesEnd – An iterator to the end of the indices

template<typename StatisticVector, typename WeightVector, typename StatisticView> static inline void setVectorToWeightedSumOfStatistics(StatisticVector &statisticVector, const WeightVector &weights, const StatisticView &statisticView)¶

Initializes a given vector by setting its element for each output to the weighted sum of the statistics in a specific view.

Template Parameters:

StatisticVector – The type of the vector to be initialized
WeightVector – The type of the vector that provides access to the weights of statistics
StatisticView – The type of the view that provides access to the statistics

Parameters:

statisticVector – A reference to an object of template type StatisticVector to be initialized
weights – A reference to an object of template type WeightVector that provides access to the weights
statisticView – A reference to an object of template type StatisticView that provides access to the statistics

class IStatisticsSubset¶

#include <statistics_subset.hpp>

Defines an interface for all classes that provide access to a subset of the statistics and allows to calculate the scores to be predicted by rules that cover such a subset.

Subclassed by AbstractStatisticsSubset< State, StatisticVector, WeightVector, IndexVector >, IResettableStatisticsSubset

Public Functions

inline virtual ~IStatisticsSubset()¶

virtual bool hasNonZeroWeight(uint32 statisticIndex) const = 0¶

Returns whether the statistics at a specific index have a non-zero weight or not.

Returns:: True, if the statistics at the given index have a non-zero weight, false otherwise

virtual void addToSubset(uint32 statisticIndex) = 0¶

Adds the statistics at a specific index to the subset in order to mark it as covered by the condition that is currently considered for refining a rule.

This function must be called repeatedly for each statistic that is covered by the current condition, immediately after the invocation of the function IWeightedStatistics::createSubset. If a rule has already been refined, each of these statistics must have been marked as covered earlier via the function IWeightedStatistics::addCoveredStatistic and must not have been marked as uncovered via the function IWeightedStatistics::removeCoveredStatistic.

This function is supposed to update any internal state of the subset that relates to the statistics that are covered by the current condition, i.e., to compute and store local information that is required by the other functions that will be called later. Any information computed by this function is expected to be reset when invoking the function resetSubset for the next time.

Parameters:: statisticIndex – The index of the covered statistic

virtual std::unique_ptr<IStatisticsUpdateCandidate> calculateScores() = 0¶

Calculates and returns the scores to be predicted by a rule that covers all statistics that have been added to the subset via the function addToSubset, as well as a numerical score that assesses the overall quality of the predicted scores.

Returns:: An unique pointer to an object of type IStatisticsUpdateCandidate that stores the scores to be predicted by the rule for each considered output, as well as a numerical score that assesses their overall quality

template<typename State, typename StatisticVector, typename WeightVector, typename IndexVector> class AbstractStatisticsSubset : public virtual IStatisticsSubset ¶

#include <statistics_subset.hpp>

An abstract base class for all classes that provide access to a subset of the statistics and allows to calculate the scores to be predicted by rules that cover such a subset.

Template Parameters:

State – The type of the state of the training process
StatisticVector – The type of the vector that is used to store the sums of statistics
WeightVector – The type of the vector that provides access to the weights of individual statistics
IndexVector – The type of the vector that provides access to the indices of the outputs that are included in the subset

Public Functions

inline AbstractStatisticsSubset(State &state, const WeightVector &weights, const IndexVector &outputIndices)¶

Parameters:

state – A reference to an object of template type State that represents the state of the training process
weights – A reference to an object of template type WeightVector that provides access to the weights of individual statistics
outputIndices – A reference to an object of template type IndexVector that provides access to the indices of the outputs that are included in the subset

inline virtual ~AbstractStatisticsSubset() override¶

inline virtual bool hasNonZeroWeight(uint32 statisticIndex) const final override¶