quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ > Class Template Reference

Calculate multiple quantiles from a variable number of elements. More...

#include <MultipleQuantiles.hpp>

Public Member Functions
	MultipleQuantilesVariableNumber (const std::size_t max_num_total, QuantilesPointer_ quantiles_ptr, const MultipleQuantilesVariableNumberOptions< Output_ > &options)
std::size_t	get_max_num_total () const
template<typename Input_ , class OutputFun_ >
void	operator() (const std::size_t num_total, Input_ *const ptr, OutputFun_ output)
template<typename Input_ , class OutputFun_ >
void	operator() (const std::size_t num_total, const std::size_t num_non_zero, Input_ *const values, OutputFun_ output)

Detailed Description

template<typename Output_, class QuantilesPointer_>
class quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ >

Calculate multiple quantiles from a variable number of elements.

Template Parameters

Output_	Floating-point type of the output quantile.
QuantilesPointer_	Pointer to a container that has a size() method and supports access by []. This may or may not be a smart pointer.

This uses the same logic as the MultipleQuantilesFixedNumber class but supports any number of elements, from zero up to a specified maximum.

Constructor & Destructor Documentation

MultipleQuantilesVariableNumber()

template<typename Output_ , class QuantilesPointer_ >

quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ >::MultipleQuantilesVariableNumber ( const std::size_t max_num_total, QuantilesPointer_ quantiles_ptr, const MultipleQuantilesVariableNumberOptions< Output_ > & options )

inline

Parameters

max_num_total	Maximum of the total number of elements. Unlike MultipleQuantilesFixedNumber, this may also be zero.
quantiles_ptr	Pointer to a container of sorted quantile probabilities. Each probability should be in \([0, 1]\).
options	Further options.

Member Function Documentation

get_max_num_total()

template<typename Output_ , class QuantilesPointer_ >

std::size_t quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ >::get_max_num_total ( ) const

inline

Returns

Maximum number of elements, as specified in the max_num_total argument of the constructor.

operator()() [1/2]

template<typename Output_ , class QuantilesPointer_ >

template<typename Input_ , class OutputFun_ >

void quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ >::operator() ( const std::size_t num_total, Input_ *const ptr, OutputFun_ output )

inline

Compute multiple quantiles from a dense array of length equal to num_total.

No consideration is given to special values like NaNs in the array. If these are to be skipped, consider using skip_values() before calling this method.

This method is not thread-safe.

Template Parameters

Input_	Numeric type of the input values.
OutputFun_	Functor that accepts a std::size_t and an Output_.

Parameters

	num_total	Total number of the elements from which to compute the quantile. This should be no greater than max_num_total.
[in]	ptr	Pointer to an array of length num_total. On output, the elements may be reordered.
	output	Function that accepts a std::size_t, the index of the probability in quantiles; and an Output_, the computed value of the quantile. This will be called once for each quantile, in order of increasing index from 0 to quantiles.size() - 1. The value of the quantile will be set to MultipleQuantilesVariableNumberOptions::placeholder if num_total == 0.

operator()() [2/2]

template<typename Output_ , class QuantilesPointer_ >

template<typename Input_ , class OutputFun_ >

void quickstats::MultipleQuantilesVariableNumber< Output_, QuantilesPointer_ >::operator() ( const std::size_t num_total, const std::size_t num_non_zero, Input_ *const values, OutputFun_ output )

inline

Overload to compute the desired quantile from a sparse vector of length num_total. This vector is assumed to have num_non_zero structural non-zeros and num_total - num_non_zero zeros.

No consideration is given to special values like NaNs in the values of the structural non-zeros. If these are to be skipped, consider using skip_values() before calling this method.

This method is not thread-safe.

Template Parameters

Input_	Numeric type of the input values.
OutputFun_	Functor that accepts a std::size_t and an Output_.

Parameters

	num_total	Total number of elements from which to compute the quantile. This should be no greater than max_num_total.
	num_non_zero	Number of elements that are structural non-zero elements. This should be no greater than num_total.
[in]	values	Pointer to the start of an array of length num_non_zero. This is expected to contain the values of the structural non-zero elements of the sparse vector. It should not contain any NaN values. On output, the elements may be reordered.
	output	Function that accepts a std::size_t, the index of the probability in quantiles; and an Output_, the computed value of the quantile. This will be called once for each quantile, in order of increasing index from 0 to quantiles.size() - 1. The value of the quantile will be set to MultipleQuantilesVariableNumberOptions::placeholder if num_total == 0.

---

The documentation for this class was generated from the following file:

• quickstats/MultipleQuantiles.hpp