holds variable type data. More...

#include <data.h>

Collaboration diagram for Dataset:

[legend]

Public Member Functions
Dataset	operator() (const vector< size_t > &idx) const
	return a slice of the data using indices idx
void	init ()
	call init at the end of constructors to define metafeatures of the data.
map< string, State >	make_features (const ArrayXXf &X, const map< string, State > &Z={}, const vector< string > &vn={}, const vector< string > &ft={})
	turns input data into a feature map
map< string, State >	copy_and_make_features (const ArrayXXf &X, const Dataset &ref_dataset, const vector< string > &vn={})
	turns input into a feature map, with feature types copied from a reference
	Dataset (std::map< string, State > &d, const Ref< const ArrayXf > &y_=ArrayXf(), bool c=false, float validation_size=0.0, float batch_size=1.0, bool shuffle_split=false)
	Dataset (const ArrayXXf &X, const Ref< const ArrayXf > &y_=ArrayXf(), const vector< string > &vn={}, const map< string, State > &Z={}, const vector< string > &ft={}, bool c=false, float validation_size=0.0, float batch_size=1.0, bool shuffle_split=false)
	Dataset (const ArrayXXf &X, const vector< string > &vn, const vector< string > &ft={}, bool c=false, float validation_size=0.0, float batch_size=1.0, bool shuffle_split=false)
	Dataset (const ArrayXXf &X, const Dataset &ref_dataset, const vector< string > &vn)
void	print () const
auto	get_X () const
Dataset	get_training_data () const
Dataset	get_validation_data () const
vector< string >	get_feature_types () const
int	get_n_samples () const
int	get_n_features () const
Dataset	get_batch () const
	select random subset of data for training weights.
float	get_batch_size ()
void	set_batch_size (float new_size)
DataType	get_feature_type (const string &name) const
std::array< Dataset, 2 >	split (const ArrayXb &mask) const
State	operator[] (std::string name) const

Public Attributes
std::vector< DataType >	unique_data_types
	keeps track of the unique data types in the dataset.
std::vector< DataType >	feature_types
	types of data in the features.
std::vector< string >	feature_names
	names of the feature types as string representations.
std::unordered_map< DataType, vector< string > >	features_of_type
	map from data types to features having that type.
std::map< string, State >	features
	dataset features, as key value pairs
ArrayXf	y
	length N array, the target label
bool	classification
	whether this is a classification problem
std::optional< std::reference_wrapper< const ArrayXXf > >	Xref
float	validation_size
	percentage of original data used for train. if 0.0, then all data is used for train and validation
bool	use_validation
bool	shuffle_split
float	batch_size
	percentage of training data size to use in each batch. if 1.0, then all data is used
bool	use_batch

Private Attributes
vector< size_t >	training_data_idx
vector< size_t >	validation_data_idx
vector< string >	feature_name_order_
	stores the original feature name order before map sorting

Detailed Description

holds variable type data.

Definition at line 50 of file data.h.

Constructor & Destructor Documentation

◆ Dataset() [1/4]

Brush::Data::Dataset::Dataset	(	std::map< string, State > &	d,
		const Ref< const ArrayXf > &	y_ = ArrayXf(),
		bool	c = false,
		float	validation_size = 0.0,
		float	batch_size = 1.0,
		bool	shuffle_split = false )

inline

initialize data from a map.

Definition at line 117 of file data.h.

◆ Dataset() [2/4]

Brush::Data::Dataset::Dataset	(	const ArrayXXf &	X,
		const Ref< const ArrayXf > &	y_ = ArrayXf(),
		const vector< string > &	vn = {},
		const map< string, State > &	Z = {},
		const vector< string > &	ft = {},
		bool	c = false,
		float	validation_size = 0.0,
		float	batch_size = 1.0,
		bool	shuffle_split = false )

inline

initialize data from a matrix with feature columns.

Definition at line 135 of file data.h.

◆ Dataset() [3/4]

Brush::Data::Dataset::Dataset	(	const ArrayXXf &	X,
		const vector< string > &	vn,
		const vector< string > &	ft = {},
		bool	c = false,
		float	validation_size = 0.0,
		float	batch_size = 1.0,
		bool	shuffle_split = false )

inline

initialize data from X and feature names

Definition at line 159 of file data.h.

◆ Dataset() [4/4]

Brush::Data::Dataset::Dataset	(	const ArrayXXf &	X,
		const Dataset &	ref_dataset,
		const vector< string > &	vn )

inline

Definition at line 181 of file data.h.

Member Function Documentation

◆ copy_and_make_features()

map< string, State > Brush::Data::Dataset::copy_and_make_features	(	const ArrayXXf &	X,
		const Dataset &	ref_dataset,
		const vector< string > &	vn = {} )

turns input into a feature map, with feature types copied from a reference

Definition at line 458 of file data.cpp.

◆ get_batch()

Dataset Brush::Data::Dataset::get_batch ( ) const

select random subset of data for training weights.

Definition at line 187 of file data.cpp.

◆ get_batch_size()

float Brush::Data::Dataset::get_batch_size ( )

Definition at line 380 of file data.cpp.

◆ get_feature_type()

DataType Brush::Data::Dataset::get_feature_type ( const string & name ) const

inline

Definition at line 238 of file data.h.

Here is the caller graph for this function:

◆ get_feature_types()

vector< string > Brush::Data::Dataset::get_feature_types ( ) const

Definition at line 217 of file data.cpp.

◆ get_n_features()

int Brush::Data::Dataset::get_n_features ( ) const

inline

Definition at line 231 of file data.h.

◆ get_n_samples()

int Brush::Data::Dataset::get_n_samples ( ) const

inline

Definition at line 225 of file data.h.

Here is the caller graph for this function:

◆ get_training_data()

Dataset Brush::Data::Dataset::get_training_data ( ) const

Definition at line 214 of file data.cpp.

◆ get_validation_data()

Dataset Brush::Data::Dataset::get_validation_data ( ) const

Definition at line 215 of file data.cpp.

◆ get_X()

auto Brush::Data::Dataset::get_X ( ) const

inline

Definition at line 212 of file data.h.

◆ init()

void Brush::Data::Dataset::init ( )

call init at the end of constructors to define metafeatures of the data.

Definition at line 250 of file data.cpp.

◆ make_features()

map< string, State > Brush::Data::Dataset::make_features	(	const ArrayXXf &	X,
		const map< string, State > &	Z = {},
		const vector< string > &	vn = {},
		const vector< string > &	ft = {} )

turns input data into a feature map

Definition at line 387 of file data.cpp.

◆ operator()()

Dataset Brush::Data::Dataset::operator() ( const vector< size_t > & idx ) const

return a slice of the data using indices idx

Definition at line 139 of file data.cpp.

◆ operator[]()

State Brush::Data::Dataset::operator[] ( std::string name ) const

inline

Definition at line 262 of file data.h.

◆ print()

void Brush::Data::Dataset::print ( ) const

inline

Definition at line 196 of file data.h.

◆ set_batch_size()

void Brush::Data::Dataset::set_batch_size ( float new_size )

Definition at line 381 of file data.cpp.

◆ split()

std::array< Dataset, 2 > Brush::Data::Dataset::split ( const ArrayXb & mask ) const

Definition at line 203 of file data.cpp.

Here is the caller graph for this function:

Member Data Documentation

◆ batch_size

float Brush::Data::Dataset::batch_size

percentage of training data size to use in each batch. if 1.0, then all data is used

Definition at line 94 of file data.h.

◆ classification

bool Brush::Data::Dataset::classification

whether this is a classification problem

Definition at line 85 of file data.h.

◆ feature_name_order_

vector<string> Brush::Data::Dataset::feature_name_order_

private

stores the original feature name order before map sorting

Definition at line 60 of file data.h.

◆ feature_names

std::vector<string> Brush::Data::Dataset::feature_names

names of the feature types as string representations.

Definition at line 70 of file data.h.

◆ feature_types

std::vector<DataType> Brush::Data::Dataset::feature_types

types of data in the features.

Definition at line 67 of file data.h.

◆ features

std::map<string, State> Brush::Data::Dataset::features

dataset features, as key value pairs

Definition at line 76 of file data.h.

◆ features_of_type

std::unordered_map<DataType,vector<string> > Brush::Data::Dataset::features_of_type

map from data types to features having that type.

Definition at line 73 of file data.h.

◆ shuffle_split

bool Brush::Data::Dataset::shuffle_split

Definition at line 91 of file data.h.

◆ training_data_idx

vector<size_t> Brush::Data::Dataset::training_data_idx

private

Definition at line 57 of file data.h.

◆ unique_data_types

std::vector<DataType> Brush::Data::Dataset::unique_data_types

keeps track of the unique data types in the dataset.

Definition at line 64 of file data.h.

◆ use_batch

bool Brush::Data::Dataset::use_batch

Definition at line 95 of file data.h.

◆ use_validation

bool Brush::Data::Dataset::use_validation

Definition at line 90 of file data.h.

◆ validation_data_idx

vector<size_t> Brush::Data::Dataset::validation_data_idx

private

Definition at line 58 of file data.h.

◆ validation_size

float Brush::Data::Dataset::validation_size

percentage of original data used for train. if 0.0, then all data is used for train and validation

Definition at line 89 of file data.h.

◆ Xref

std::optional<std::reference_wrapper<const ArrayXXf> > Brush::Data::Dataset::Xref

Definition at line 86 of file data.h.

◆ y

ArrayXf Brush::Data::Dataset::y

length N array, the target label

Definition at line 82 of file data.h.

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

Public Member Functions

Public Attributes

Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Dataset() [1/4]

◆ Dataset() [2/4]

◆ Dataset() [3/4]

◆ Dataset() [4/4]

Member Function Documentation

◆ copy_and_make_features()

◆ get_batch()

◆ get_batch_size()

◆ get_feature_type()

◆ get_feature_types()

◆ get_n_features()

◆ get_n_samples()

◆ get_training_data()

◆ get_validation_data()

◆ get_X()

◆ init()

◆ make_features()

◆ operator()()

◆ operator[]()

◆ print()

◆ set_batch_size()

◆ split()

Member Data Documentation

◆ batch_size

◆ classification

◆ feature_name_order_

◆ feature_names

◆ feature_types

◆ features

◆ features_of_type

◆ shuffle_split

◆ training_data_idx

◆ unique_data_types

◆ use_batch

◆ use_validation

◆ validation_data_idx

◆ validation_size

◆ Xref

◆ y