#include <ml/ml_data/column_statistics.hpp>

Public Member Functions
	column_statistics (std::string column_name, ml_column_mode mode, flex_type_enum original_column_type)

size_t	num_observations () const

void	initialize ()
	Initialize the statistics – counting, mean, and stdev.

void	update_categorical_statistics (size_t thread_idx, const std::vector< size_t > &cat_index_vect) GL_HOT
	Update categorical statistics for a batch of categorical indices.

void	update_numeric_statistics (size_t thread_idx, const std::vector< double > &value_vect) GL_HOT
	Update categorical statistics for a batch of real values.

void	update_dict_statistics (size_t thread_idx, const std::vector< std::pair< size_t, double > > &dict) GL_HOT
	Update statistics after observing a dictionary.

void	finalize ()

void	reindex (const std::vector< size_t > &new_index_map, size_t new_column_size)

void	merge_in (const column_statistics &other)

size_t	get_version () const

void	save_impl (turi::oarchive &oarc) const

void	load_version (turi::iarchive &iarc, size_t version)

bool	is_equal (const column_statistics *other_ptr) const

bool	operator== (const column_statistics &other) const

bool	operator!= (const column_statistics &other) const

Detailed Description

column_metadata contains "meta data" concerning indexing of a single column of an SFrame. A collection of meta_data column objects is "all" the metadata required in the ml_data container.

Definition at line 35 of file column_statistics.hpp.

Constructor & Destructor Documentation

◆ column_statistics()

turi::ml_data_internal::column_statistics::column_statistics	(	std::string	column_name,
		ml_column_mode	mode,
		flex_type_enum	original_column_type
	)

Default constructor.

Member Function Documentation

◆ finalize()

void turi::ml_data_internal::column_statistics::finalize ( )

Perform final computations on the different statistics. Must be called after all the data is filled.

◆ get_version()

size_t turi::ml_data_internal::column_statistics::get_version ( ) const

inline

Returns the current serialization version of this model.

Definition at line 351 of file column_statistics.hpp.

◆ is_equal()

bool turi::ml_data_internal::column_statistics::is_equal ( const column_statistics * other_ptr ) const

For debugging purposes.

◆ load_version()

void turi::ml_data_internal::column_statistics::load_version	(	turi::iarchive &	iarc,
		size_t	version
	)

Load the object.

◆ merge_in()

void turi::ml_data_internal::column_statistics::merge_in ( const column_statistics & other )

Merges in statistics from another column_statistics object.

◆ num_observations()

size_t turi::ml_data_internal::column_statistics::num_observations ( ) const

inline

Returns the number of seen by the methods collecting the statistics.

Definition at line 54 of file column_statistics.hpp.

◆ operator!=()

bool turi::ml_data_internal::column_statistics::operator!= ( const column_statistics & other ) const

Inequality testing – slow! Use for debugging/testing

◆ operator==()

bool turi::ml_data_internal::column_statistics::operator== ( const column_statistics & other ) const

Equality testing – slow! Use for debugging/testing

◆ reindex()

void turi::ml_data_internal::column_statistics::reindex	(	const std::vector< size_t > &	new_index_map,
		size_t	new_column_size
	)

Reindex, typically according to a global map of things.

◆ save_impl()

void turi::ml_data_internal::column_statistics::save_impl ( turi::oarchive & oarc ) const

Serialize the object (save).

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

ml/ml_data/column_statistics.hpp

Public Member Functions