Data container. More...

#include <CosanData.h>

Inheritance diagram for Cosan::CosanData< NumericType >:

Public Member Functions
	CosanData ()=default

	CosanData (gsl::index nrows, gsl::index ncols, NumericType lb=0, NumericType ub=1)
	Generate random matrix with each entry uniformly sampled from lb to lb. Dimension is nrows by ncols. More...

	CosanData (const CosanMatrix< NumericType > &inputX)
	Get CosanData from CosanMatrix<NUmericType> inputX. More...

	CosanData (const CosanMatrix< NumericType > &inputX, const CosanMatrix< NumericType > &inputY)
	Get CosanData from CosanMatrix<NUmericType> inputX, inputY. More...

	CosanData (const std::vector< NumericType > &inputX, gsl::index nrows, const std::string &order="rowfirst")
	Get CosanData from std::vector inputX, fill the data either by 'rowfirst' or 'columnfirst'. More...

	CosanData (const std::vector< NumericType > &inputX, const std::vector< NumericType > &inputY, gsl::index nrows, const std::string &order="rowfirst")
	Get CosanData from std::vector inputX and inputY, fill the data either by 'rowfirst' or 'columnfirst'. More...

virtual const std::string	GetName () const
	Get the name of the object. More...

Public Member Functions inherited from Cosan::CosanRawData< NumericType >
	CosanRawData ()=default

	CosanRawData (const std::string &srcX, const std::string &srcY)
	Constructor: Read data X and Y from csv files and form raw data container. More...

	CosanRawData (const std::string &srcX)
	Constructor: Read data X from csv and form raw data container. More...

void	SetInput (const std::string &srcX)
	Update input X from csv file. More...

void	SetTarget (const std::string &srcY)
	Update target Y from csv file. More...

void	ConcatenateData (const CosanMatrix< NumericType > &inputX)
	Concatenate X using CosanMatrix<NumericType> input X. Add new columns. More...

void	UpdateData (const CosanMatrix< NumericType > &inputX)
	Update X using CosanMatrix<NumericType> input X. More...

void	UpdateData (const CosanMatrix< NumericType > &inputX, const CosanMatrix< NumericType > &inputY)
	Update X and Y using CosanMatrix<NumericType> inputX,inputY. More...

void	UpdateCat (const std::vector< std::string > &inputX)
	Update categorical vector svaluesX using std::vector<std::string> & inputX. More...

void	UpdateCat (const std::vector< std::string > &inputX, const std::vector< std::string > &inputY)
	Update categorical vector svaluesX,svaluesY using std::vector<std::string> & inputX,inputY. More...

CosanMatrix< NumericType >	GetInput ()
	Get a copy of CosanMatrix<NumericType> X. More...

CosanMatrix< NumericType >	GetTarget ()
	Get a copy of CosanMatrix<NumericType> Y. More...

const CosanMatrix< NumericType > &	GetInput () const
	Get a const reference to const CosanMatrix<NumericType> X. More...

const CosanMatrix< NumericType > &	GetTarget () const
	Get a const reference to const CosanMatrix<NumericType> Y. More...

std::tuple< gsl::index, gsl::index >	GetMissingNumber ()
	Get the total number data information. More...

const std::string &	GetSummaryMessageX () const
	Get the summary message on reading csv file on X. More...

const std::string &	GetSummaryMessageY () const
	Get the summary message on reading csv file on Y. More...

std::unordered_map< gsl::index, gsl::index > &	GetRawToNumIdx ()
	Raw data column index to numeric data matrix X column index. More...

std::unordered_map< gsl::index, gsl::index > &	GetRawToCatIdx ()
	Raw data column index to categorical data column index. More...

std::vector< std::vector< gsl::index > >	GetIdxpinfX () const
	Get the position of positive infinity in the origin data X. More...

std::vector< std::vector< gsl::index > >	GetIdxminfX () const
	Get the position of negative infinity in the origin data X. More...

std::vector< std::vector< gsl::index > >	GetIdxmissingX () const
	Get the position of missing in the origin data X. More...

std::vector< std::vector< gsl::index > >	GetIdxpinfY () const
	Get the position of positive infinity in the origin data Y. More...

std::vector< std::vector< gsl::index > >	GetIdxminfY () const
	Get the position of negative infinity in the origin data Y. More...

std::vector< std::vector< gsl::index > >	GetIdxmissingY () const
	Get the position of missing in the origin data Y. More...

std::set< gsl::index >	GetcolCatX () const
	Get the column index (in the origin X of csv file) where the column is of categorical type. More...

std::set< gsl::index >	GetcolCatY () const
	Get the column index (in the origin Y of csv file) where the column is of categorical type. More...

bool	GetcatY () const
	True if Y is categorical data type. False otherwise. More...

gsl::index	GetrowsX ()
	Get the number of rows for X. More...

gsl::index	GetrowsY ()
	Get the number of rows for Y. More...

gsl::index	GetcolsX ()
	Get the number of columns for X. More...

gsl::index	GetcolsY ()
	Get the number of columns for Y. More...

std::vector< std::string >	GetsvaluesX () const
	Get the vector of categorical data from X. order: row first. More...

std::vector< std::string >	GetsvaluesY () const
	Get the vector of categorical data from Y. order: row first. More...

CosanMatrix< NumericType >	GetType ()

Public Member Functions inherited from Cosan::CosanBO
	CosanBO ()
	Default constructor. More...

Additional Inherited Members
Protected Attributes inherited from Cosan::CosanRawData< NumericType >
CosanMatrix< NumericType >	X
	Numeric data from origin CSV file for X. More...

CosanMatrix< NumericType >	Y
	Numeric data from origin CSV file for Y. More...

CosanMatrix< NumericType >	__TYPE

std::string	SummaryMessageX
	Loading message. More...

std::string	SummaryMessageY

std::vector< std::vector< gsl::index > >	IdxpinfX
	position for positive, negative infinity and missing values. More...

std::vector< std::vector< gsl::index > >	IdxminfX

std::vector< std::vector< gsl::index > >	IdxmissingX

std::vector< std::vector< gsl::index > >	IdxpinfY

std::vector< std::vector< gsl::index > >	IdxminfY

std::vector< std::vector< gsl::index > >	IdxmissingY

std::set< gsl::index >	colCatX
	column idx in the origin data that is categorical data. More...

std::set< gsl::index >	colCatY

bool	catY = false
	true mean respone variable Y is categorical data. More...

gsl::index	rowsX = 0
	number of rows. More...

gsl::index	colsX = 0

gsl::index	rowsY = 0
	number of columns More...

gsl::index	colsY = 0

std::vector< std::string >	svaluesX

std::vector< std::string >	svaluesY

Detailed Description

template<Numeric NumericType>
class Cosan::CosanData< NumericType >

Data container.

Every constructor needs to have at least one input. To obtain CosanData, the following constructors can be used:

CosanData(gsl::index nrows,gsl::index ncols,NumericType lb=0,NumericType ub = 1)

CosanData(const CosanMatrix<NumericType> & inputX)

CosanData(const CosanMatrix<NumericType>& inputX,const CosanMatrix<NumericType>& inputY)

CosanData(const std::vector<NumericType>& inputX,gsl::index nrows,const std::string & order = "rowfirst")

CosanData(const std::vector<NumericType>& inputX,const std::vector<NumericType>& inputY,gsl::index nrows,const std::string & order = "rowfirst")

Definition at line 546 of file CosanData.h.

Constructor & Destructor Documentation

◆ CosanData() [1/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData ( )

default

◆ CosanData() [2/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData	(	gsl::index	nrows,
		gsl::index	ncols,
		NumericType	lb = `0`,
		NumericType	ub = `1`
	)

inline

Generate random matrix with each entry uniformly sampled from lb to lb. Dimension is nrows by ncols.

Definition at line 552 of file CosanData.h.

                                                                                         :CosanRawData<NumericType>(){
                 this->X.resize(nrows,ncols);
                 std::default_random_engine generator;
                 std::uniform_real_distribution<double> distribution(lb,ub);
                 for (gsl::index  i = 0;i<nrows*ncols;i++){
                     this->X(i/ncols,i%ncols) =distribution(generator);
                 }
             }

◆ CosanData() [3/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData ( const CosanMatrix< NumericType > & inputX )

inline

Get CosanData from CosanMatrix<NUmericType> inputX.

Definition at line 563 of file CosanData.h.

                                                               :CosanRawData<NumericType>(){
                 static_assert(std::is_arithmetic<NumericType>::value, "NumericType must be numeric");
                 this->X = inputX;
             }

◆ CosanData() [4/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData	(	const CosanMatrix< NumericType > &	inputX,
		const CosanMatrix< NumericType > &	inputY
	)

inline

Get CosanData from CosanMatrix<NUmericType> inputX, inputY.

Definition at line 570 of file CosanData.h.

                                                                                                      :CosanRawData<NumericType>(){
                 static_assert(std::is_arithmetic<NumericType>::value, "NumericType must be numeric");
                 this->X = inputX;
                 this->Y = inputY;}

◆ CosanData() [5/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData	(	const std::vector< NumericType > &	inputX,
		gsl::index	nrows,
		const std::string &	order = `"rowfirst"`
	)

inline

Get CosanData from std::vector inputX, fill the data either by 'rowfirst' or 'columnfirst'.

Definition at line 577 of file CosanData.h.

                                                                                                                 :CosanRawData<NumericType>(){
                 if (nrows>inputX.size() || inputX.size()%nrows!=0){
                     throw std::invalid_argument(
                             fmt::format("Incorrect nrows specification, should be less than or equal to input vector size and size is divisible by nrows. Input vector size is "
                                         "{:} and nrows is {:}",inputX.size(),nrows));
                 }
                 this->X.resize(nrows,inputX.size()/nrows);
                 gsl::index i =0,__cols=inputX.size()/nrows;
                 if (order=="columnfirst"){
                     i =0;
                     for (auto &each :inputX ){
                         this->X(i%nrows,i/nrows) = each;
                         i++;
                     }
                     return;
                 }
                 i =0;
                 for (auto &each :inputX ){
                     this->X(i/__cols,i%__cols) = each;
                     i++;
                 }
                 return;
             }

◆ CosanData() [6/6]

template<Numeric NumericType>

Cosan::CosanData< NumericType >::CosanData	(	const std::vector< NumericType > &	inputX,
		const std::vector< NumericType > &	inputY,
		gsl::index	nrows,
		const std::string &	order = `"rowfirst"`
	)

inline

Get CosanData from std::vector inputX and inputY, fill the data either by 'rowfirst' or 'columnfirst'.

Definition at line 605 of file CosanData.h.

                                                                                                                                                       :CosanRawData<NumericType>(){
                 if (nrows>inputX.size() || inputX.size()%nrows!=0 || nrows!=inputY.size()){
                     throw std::invalid_argument(
                             fmt::format("Incorrect nrows specification, should be less than or equal to input vector size and size is divisible by nrows. inputY size should also be equal to nrows."
                                         "inputX vector size is {:}, inputY vector size is {:} and nrows is {:}",inputX.size(),inputY.size(),nrows));
                 }
                 this->X.resize(nrows,inputX.size()/nrows);
                 this->Y.resize(nrows,1);
                 gsl::index i =0,__cols=inputX.size()/nrows;
                 for (auto &each:inputY){
                     this->Y(i,0) = each;
                     i++;
                 }
                 if (order=="columnfirst"){
                     i =0;
                     for (auto &each :inputX ){
                         this->X(i%nrows,i/nrows) = each;
                         i++;
                     }
                     return;
                 }
                 i =0;
                 for (auto &each :inputX ){
                     this->X(i/__cols,i%__cols) = each;
                     i++;
                 }
                 return;
             }

Member Function Documentation

◆ GetName()

template<Numeric NumericType>

virtual const std::string Cosan::CosanData< NumericType >::GetName ( ) const

inlinevirtual

Get the name of the object.

Reimplemented from Cosan::CosanRawData< NumericType >.

Definition at line 637 of file CosanData.h.

637 {return "Processed Data Object.";}

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

cosan/data/CosanData.h

Public Member Functions

Additional Inherited Members

Detailed Description

template<Numeric NumericType> class Cosan::CosanData< NumericType >

Constructor & Destructor Documentation

◆ CosanData() [1/6]

◆ CosanData() [2/6]

◆ CosanData() [3/6]

◆ CosanData() [4/6]

◆ CosanData() [5/6]

◆ CosanData() [6/6]

Member Function Documentation

◆ GetName()

template<Numeric NumericType>
class Cosan::CosanData< NumericType >