A data structure for holding labeled columns of data, where each column is a vector of real numbers. More...

#include <data_table.h>

Public Member Functions
	data_table ()
	Default constructor.

	data_table (const std::map< std::string, vec< real > > &table)
	Construct from map of column vectors.

	data_table (size_t num_rows, const std::vector< std::string > &column_names)
	Construct a data table with preallocated size.

	data_table (const data_table &other)
	Copy constructor from another data table.

	data_table (data_table &&other) noexcept
	Move constructor from another data table, leaving the other empty.

size_t	rows () const
	Get the (maximum) number of rows in the data table.

size_t	cols () const
	Get the number of columns in the data table.

size_t	size () const
	Get the total number of elements in the data table, i.e.

bool	empty () const
	Check whether the data table is empty (i.e.

void	clear ()
	Remove all columns from the data table, leaving it empty.

std::vector< std::string >	header () const
	Get a list of the column names in the data table.

bool	has_column (const std::string &name) const
	Check whether the data table has a column with the given name.

std::vector< vec< real > > &	data ()
	Get the columns of the data table as a simple vector of column vectors, without the column names or indices, by reference.

const std::vector< vec< real > > &	data () const
	Get the columns of the data table as a simple vector of column vectors, without the column names or indices.

vec< real > &	operator[] (const std::string &name)
	Access a column by name, returning a reference to the column vector.

const vec< real > &	operator[] (const std::string &name) const
	Access a column by name, returning a const reference to the column vector.

vec< real > &	at (const std::string &name)
	Access a column by name, returning a reference to the column vector.

vec< real > &	operator[] (size_t idx)
	Access a column by index, returning a reference to the column vector.

data_table	select (const std::vector< std::string > &cols) const
	Select a subset of columns from the table, returning a new table containing only the selected columns.

real &	at (const std::string &col, size_t row)
	Access an element by row index and column name, returning a reference to the value.

const real &	at (const std::string &col, size_t row) const
	Access an element by row index and column name, returning a const reference to the value.

std::unordered_map< std::string, real >	row (size_t idx) const
	Get an entire row as a map of column names to values.

vec< real >	row_vec (size_t idx) const
	Get an entire row as a vector of values, in the same order as the columns in the data table.

data_table	head (size_t n=5) const
	Get the first n rows of the data table as a new data table.

data_table	tail (size_t n=5) const
	Get the last n rows of the data table as a new data table.

data_table &	insert (const std::string &name, const vec< real > &data)
	Insert a new column into the data table with the given name and data.

data_table &	insert (const std::string &name, size_t num_rows, real value=0.0)
	Insert a new column into the data table with the given name, number of rows, and constant value.

data_table &	drop_column (const std::string &name)
	Drop a column from the data table by name.

data_table &	drop_columns (const std::vector< std::string > &names)
	Drop multiple columns from the data table by name.

data_table &	rename (const std::string &old_name, const std::string &new_name)
	Rename a column in the data table from old_name to new_name.

mat< real >	to_matrix () const
	Convert the data table to a matrix, where each column of the matrix corresponds to a column in the data table, and each row corresponds to a row in the data table.

data_table &	from_matrix (const mat< real > &m, const std::vector< std::string > &col_names)
	Convert a matrix to a data table, where each column of the matrix corresponds to a column in the data table, and each row corresponds to a row in the data table.

std::string	to_string (unsigned int max_rows=8, unsigned int precision=6, unsigned int max_width=12) const
	Convert the data table to string representation.

	operator std::string ()
	Convert the data table to string representation.

Friends
std::ostream &	operator<< (std::ostream &out, const data_table &obj)
	Stream the data table in string representation to an output stream (std::ostream)

Detailed Description

A data structure for holding labeled columns of data, where each column is a vector of real numbers.

Constructor & Destructor Documentation

◆ data_table() [1/2]

theoretica::data_table::data_table ( const std::map< std::string, vec< real > > & table )

inline

Construct from map of column vectors.

The order of columns is determined by the order of the map iteration.

Parameters

table A map where keys are column names and values are column vectors.

◆ data_table() [2/2]

theoretica::data_table::data_table	(	size_t	num_rows,
		const std::vector< std::string > &	column_names
	)

inline

Construct a data table with preallocated size.

Parameters

num_rows	The number of rows in the data table
column_names	The names of the columns in the data table

Member Function Documentation

◆ at() [1/3]

real & theoretica::data_table::at	(	const std::string &	col,
		size_t	row
	)

inline

Access an element by row index and column name, returning a reference to the value.

Throws an out_of_range exception if the column name is not found or the row index is out of range.

Parameters

row	The index of the row to access.
col	The name of the column to access.

Returns: A reference to the value at the given row and column.

◆ at() [2/3]

const real & theoretica::data_table::at	(	const std::string &	col,
		size_t	row
	)		const

inline

Access an element by row index and column name, returning a const reference to the value.

Throws an out_of_range exception if the column name is not found or the row index is out of range.

Parameters

row	The index of the row to access.
col	The name of the column to access.

Returns: A const reference to the value at the given row and column.

◆ at() [3/3]

vec< real > & theoretica::data_table::at ( const std::string & name )

inline

Access a column by name, returning a reference to the column vector.

Throws an out_of_range exception if the column name is not found.

Parameters

name	The name of the column to access.

Returns: A reference to the column vector with the given name.

◆ cols()

size_t theoretica::data_table::cols ( ) const

inline

Get the number of columns in the data table.

Returns: The number of columns in the data table.

◆ data() [1/2]

std::vector< vec< real > > & theoretica::data_table::data ( )

inline

Get the columns of the data table as a simple vector of column vectors, without the column names or indices, by reference.

Returns: A vector of column vectors in the data table.

◆ data() [2/2]

const std::vector< vec< real > > & theoretica::data_table::data ( ) const

inline

Get the columns of the data table as a simple vector of column vectors, without the column names or indices.

Returns: A vector of column vectors in the data table.

◆ drop_column()

data_table & theoretica::data_table::drop_column ( const std::string & name )

inline

Drop a column from the data table by name.

If the column name is not found, the table is unchanged. This operation is O(K) in the number of columns K, since it requires updating the indices of all subsequent columns.

Parameters

name	The name of the column to drop.

◆ drop_columns()

data_table & theoretica::data_table::drop_columns ( const std::vector< std::string > & names )

inline

Drop multiple columns from the data table by name.

Column names that are not found are ignored. This operation is O(K) with respect to the number of columns K, since it requires updating the indices of all subsequent columns for each dropped column.

Parameters

names The names of the columns to drop.

◆ empty()

bool theoretica::data_table::empty ( ) const

inline

Check whether the data table is empty (i.e.

has no columns or all columns are empty).

Returns: True if the data table is empty, false otherwise.

◆ from_matrix()

data_table & theoretica::data_table::from_matrix	(	const mat< real > &	m,
		const std::vector< std::string > &	col_names
	)

inline

Convert a matrix to a data table, where each column of the matrix corresponds to a column in the data table, and each row corresponds to a row in the data table.

The column names are taken from the provided vector, and if there are fewer column names than columns in the matrix, the remaining columns are named "col_i" where i is the column index. If there are more column names than columns in the matrix, the extra column names are ignored.

Parameters

m	The matrix to read elements from.
col_names	Names of the columns in the resulting data table.

◆ has_column()

bool theoretica::data_table::has_column ( const std::string & name ) const

inline

Check whether the data table has a column with the given name.

Parameters

name	The name of the column to check for.

Returns: Whether the table contains a column with the given name.

◆ head()

data_table theoretica::data_table::head ( size_t n = 5 ) const

inline

Get the first n rows of the data table as a new data table.

If n is greater than the number of rows in the table, the entire table is returned.

Parameters

n	The number of rows to include in the head of the data table.

Returns: A new data table containing the first n rows of the original table.

◆ header()

std::vector< std::string > theoretica::data_table::header ( ) const

inline

Get a list of the column names in the data table.

Returns: A vector of column names in the data table.

◆ insert() [1/2]

data_table & theoretica::data_table::insert	(	const std::string &	name,
		const vec< real > &	data
	)

inline

Insert a new column into the data table with the given name and data.

If a column with the same name already exists, it is overwritten.

Parameters

name	The name of the column to add.
data	The data representing the new column.

◆ insert() [2/2]

data_table & theoretica::data_table::insert	(	const std::string &	name,
		size_t	num_rows,
		real	value = `0.0`
	)

inline

Insert a new column into the data table with the given name, number of rows, and constant value.

If a column with the same name already exists, it is overwritten.

Parameters

name	The name of the column to add.
num_rows	The number of rows in the new column.
value	The value to fill the new column with.

◆ operator[]() [1/3]

vec< real > & theoretica::data_table::operator[] ( const std::string & name )

inline

Access a column by name, returning a reference to the column vector.

Throws an out_of_range exception if the column name is not found.

Parameters

name	The name of the column to access.

Returns: A reference to the column vector with the given name.

◆ operator[]() [2/3]

const vec< real > & theoretica::data_table::operator[] ( const std::string & name ) const

inline

Access a column by name, returning a const reference to the column vector.

Throws an out_of_range exception if the column name is not found.

Parameters

name	The name of the column to access.

Returns: A const reference to the column vector with the given name.

◆ operator[]() [3/3]

vec< real > & theoretica::data_table::operator[] ( size_t idx )

inline

Access a column by index, returning a reference to the column vector.

Does not perform bounds checking, so the behavior is undefined if the index is out of range, for safer access consider using the at() method instead, which performs bounds checking.

Parameters

idx	The index of the column to access.

Returns: A reference to the column vector at the given index.

◆ rename()

data_table & theoretica::data_table::rename	(	const std::string &	old_name,
		const std::string &	new_name
	)

inline

Rename a column in the data table from old_name to new_name.

If old_name is not found, the table is unchanged, while if new_name already exists, it is overwritten.

Parameters

old_name	The old name of the column to rename
new_name	The new name of the column

◆ row()

std::unordered_map< std::string, real > theoretica::data_table::row ( size_t idx ) const

inline

Get an entire row as a map of column names to values.

Missing values for columns that do not have enough rows are filled with NaN.

Parameters

idx	The index of the row to access.

Returns: A map of column names to values for the given row.

◆ row_vec()

vec< real > theoretica::data_table::row_vec ( size_t idx ) const

inline

Get an entire row as a vector of values, in the same order as the columns in the data table.

Missing values for columns that do not have enough rows are filled with NaN.

Parameters

idx	The index of the row to access.

Returns: A map of column names to values for the given row.

◆ rows()

size_t theoretica::data_table::rows ( ) const

inline

Get the (maximum) number of rows in the data table.

This operation is O(K) in the number of columns K.

Returns: The size of the first column or 0 if the table is empty.

◆ select()

data_table theoretica::data_table::select ( const std::vector< std::string > & cols ) const

inline

Select a subset of columns from the table, returning a new table containing only the selected columns.

If a column name in the selection list is not found in the data table, it is ignored.

Note: The current implementation immediately returns a new data table, but in the future it may be optimized to return a view of the original data table instead, to avoid unnecessary copying.

Parameters

cols	A vector of column names to select from the data table.

Returns: A new data table containing only the selected columns.

◆ size()

size_t theoretica::data_table::size ( ) const

inline

Get the total number of elements in the data table, i.e.

N. rows * N. columns.

Returns: The total number of elements in the data table.

◆ tail()

data_table theoretica::data_table::tail ( size_t n = 5 ) const

inline

Get the last n rows of the data table as a new data table.

If n is greater than the number of rows in the table, the entire table is returned.

Parameters

n	The number of rows to include in the tail of the data table.

Returns: A new data table containing the last n rows of the original table.

◆ to_matrix()

mat< real > theoretica::data_table::to_matrix ( ) const

inline

Convert the data table to a matrix, where each column of the matrix corresponds to a column in the data table, and each row corresponds to a row in the data table.

Missing values are filled with NaN.

Returns: A matrix representation of the data table.

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

io/data_table.h

Public Member Functions

Friends

Detailed Description

Constructor & Destructor Documentation

◆ data_table() [1/2]

◆ data_table() [2/2]

Member Function Documentation

◆ at() [1/3]

◆ at() [2/3]

◆ at() [3/3]

◆ cols()

◆ data() [1/2]

◆ data() [2/2]

◆ drop_column()

◆ drop_columns()

◆ empty()

◆ from_matrix()

◆ has_column()

◆ head()

◆ header()

◆ insert() [1/2]

◆ insert() [2/2]

◆ operator[]() [1/3]

◆ operator[]() [2/3]

◆ operator[]() [3/3]

◆ rename()

◆ row()

◆ row_vec()

◆ rows()

◆ select()

◆ size()

◆ tail()

◆ to_matrix()