khiops.sklearn.dataset

Submodule of khiops.sklearn

Classes for handling diverse data tables

CLASS	DESCRIPTION
`Dataset`	A representation of a dataset
`DatasetTable`	A generic dataset table
`PandasTable`	DatasetTable encapsulating a pandas dataframe
`NumpyTable`	DatasetTable encapsulating a NumPy array
`SparseTable`	DatasetTable encapsulating a SciPy sparse matrix

FUNCTION	DESCRIPTION
`check_dataset_spec`	Checks that a dataset spec is valid
`table_name_of_path`	Returns the table name as the last fragment of the table data path
`get_khiops_type`	Translates a numpy dtype to a Khiops dictionary type
`get_khiops_variable_name`	Return the khiops variable name associated to a column id
`read_internal_data_table`	Reads into a DataFrame a data table file with the internal format settings
`write_internal_data_table`	Writes a DataFrame to data table file with the internal format settings

Classes

`Dataset(X, y=None, categorical_target=True)`

A representation of a dataset

PARAMETER	DESCRIPTION
`X`	Either: A single dataframe A `dict` dataset specification TYPE: `DataFrame or dict`
`y`	The target column. TYPE: `(Series, DataFrame or ndarray)` DEFAULT: `None`
`categorical_target`	`True` if the vector `y` should be considered as a categorical variable. If `False` it is considered as numeric. Ignored if `y` is `None`. TYPE: `bool` DEFAULT: `True`

METHOD	DESCRIPTION
`to_spec`	Returns a dictionary specification of this dataset
`copy`	Creates a copy of the dataset
`get_table`	Returns a table by its name
`create_khiops_dictionary_domain`	Creates a Khiops dictionary domain representing this dataset
`create_table_files_for_khiops`	Prepares the tables of the dataset to be used by Khiops

ATTRIBUTE	DESCRIPTION
`table_type`	type : The table type of this dataset's tables
`is_multitable`	bool : `True` if the dataset is multitable

Attributes

`table_type` `property`

type : The table type of this dataset's tables

Possible values:

`is_multitable` `property`

bool : True if the dataset is multitable

Methods:

`to_spec()`

Returns a dictionary specification of this dataset

`copy()`

Creates a copy of the dataset

Referenced pandas.DataFrame, numpy.ndarray and scipy.sparse.spmatrix objects in tables are copied as references.

`get_table(table_name)`

Returns a table by its name

PARAMETER	DESCRIPTION
`table_name`	The name of the table to be retrieved.

RETURNS	DESCRIPTION
`DatasetTable`	The table object for the specified name.

RAISES	DESCRIPTION
`KeyError`	If there is no table with the specified name.

`create_khiops_dictionary_domain()`

Creates a Khiops dictionary domain representing this dataset

RETURNS	DESCRIPTION
`DictionaryDomain`	The DictionaryDomain object representing this dataset

`create_table_files_for_khiops(output_dir, sort=True)`

Prepares the tables of the dataset to be used by Khiops

If this is a multi-table dataset it will create sorted copies the tables.

PARAMETER	DESCRIPTION
`output_dir`	The directory where the sorted tables will be created. TYPE: `str`

RETURNS	DESCRIPTION
`tuple`	A tuple containing: The path of the main table A dictionary containing the relation [data-path -> file-path] for the secondary tables. The dictionary is empty for monotable datasets.

`DatasetTable(name, key=None)`

Bases: ABC

A generic dataset table

METHOD	DESCRIPTION
`check_key`	Checks that the key columns exist
`create_table_file_for_khiops`	Creates a copy of the table at the specified directory
`n_features`	Returns the number of features of the table
`create_khiops_dictionary`	Creates a Khiops dictionary representing this table

Methods:

`check_key()`

Checks that the key columns exist

`create_table_file_for_khiops(output_dir, sort=True)` `abstractmethod`

Creates a copy of the table at the specified directory

`n_features()`

Returns the number of features of the table

The target column does not count.

`create_khiops_dictionary()`

Creates a Khiops dictionary representing this table

RETURNS	DESCRIPTION
`Dictionary`	The Khiops Dictionary object describing this table's schema

`PandasTable(name, dataframe, key=None)`

Bases: DatasetTable

DatasetTable encapsulating a pandas dataframe

PARAMETER	DESCRIPTION
`name`	Name for the table. TYPE: `str`
`dataframe`	The data frame to be encapsulated. It must be non-empty. TYPE: `DataFrame`
`key`	The names of the columns composing the key. TYPE: `list of str` DEFAULT: `None`

`NumpyTable(name, array, key=None)`

Bases: DatasetTable

DatasetTable encapsulating a NumPy array

PARAMETER	DESCRIPTION
`name`	Name for the table. TYPE: `str`
`array`	The data frame to be encapsulated. TYPE: `numpy.ndarray of shape (n_samples, n_features_in) or Sequence`
`key`	The names of the columns composing the key. TYPE: `array-like of int` DEFAULT: `None`

`SparseTable(name, matrix, key=None)`

Bases: DatasetTable

DatasetTable encapsulating a SciPy sparse matrix

PARAMETER	DESCRIPTION
`name`	Name for the table. TYPE: `str`
`matrix`	The sparse matrix to be encapsulated. TYPE: `spmatrix`
`key`	The names of the columns composing the key. TYPE: `list of str` DEFAULT: `None`

METHOD	DESCRIPTION
`create_khiops_dictionary`	Creates a Khiops dictionary representing this sparse table

Methods:

`create_khiops_dictionary()`

Creates a Khiops dictionary representing this sparse table

Adds metadata to each sparse variable

RETURNS	DESCRIPTION
`Dictionary`	The Khiops [Dictionary][khiops.dictionary.Dictionary] object describing this table's schema

Functions:

`check_dataset_spec(ds_spec)`

Checks that a dataset spec is valid

PARAMETER	DESCRIPTION
`ds_spec`	A specification of a multi-table dataset (see Multi-Table Learning Primer). TYPE: `dict`

RAISES	DESCRIPTION
`TypeError`	If there are objects of the spec with invalid type.
`ValueError`	If there are objects of the spec with invalid values.

`table_name_of_path(table_path)`

Returns the table name as the last fragment of the table data path

PARAMETER	DESCRIPTION
`table_path`	Data path of the table, in the format "path/to/table".

RETURNS	DESCRIPTION
`str`	The name of the table.

`get_khiops_type(numpy_type, categorical_str_max_size=None)`

Translates a numpy dtype to a Khiops dictionary type

PARAMETER	DESCRIPTION
`numpy_type`	Numpy type of the column TYPE: `dtype`
`categorical_str_max_size`	Maximum length of the entries of the column whose type is `numpy_type`. TYPE: `int` DEFAULT: `None`

RETURNS	DESCRIPTION
`str`	Khiops type name. Either "Categorical", "Text", "Numerical" or "Timestamp".

Note

The "Text" Khiops type is inferred if the NumPy type is "string"
and the maximum length of the entries of that type is greater than 100.

`get_khiops_variable_name(column_id)`

Return the khiops variable name associated to a column id

`read_internal_data_table(file_path_or_stream, column_dtypes=None)`

Reads into a DataFrame a data table file with the internal format settings

The table is read with the following settings:

Use tab as separator
Read the column names from the first line
Use '"' as quote character
Use csv.QUOTE_MINIMAL
double quoting enabled (quotes within quotes can be escaped with '""')
UTF-8 encoding
User-specified dtypes (optional)

PARAMETER	DESCRIPTION
`file_path_or_stream`	The path of the internal data table file to be read or a readable file object. TYPE: `str or file object`
`column_dtypes`	Dictionary linking column names with dtypes. See `dtype` parameter of the pandas.read_csv function. If not set, then the column types are detected automatically by pandas. TYPE: `dict` DEFAULT: `None`

RETURNS	DESCRIPTION
`DataFrame`	The dataframe representation of the data table.

`write_internal_data_table(dataframe, file_path_or_stream)`

Writes a DataFrame to data table file with the internal format settings

The table is written with the following settings:

Use tab as separator
Write the column names on the first line
Use '"' as quote character
Use csv.QUOTE_MINIMAL
double quoting enabled (quotes within quotes can be escaped with '""')
UTF-8 encoding
The index is not written

Khiops cannot handle multi-line records. Hence, the carriage returns / line feeds need to be removed from the records before data is handed over to Khiops.

PARAMETER	DESCRIPTION
`dataframe`	The dataframe to write. TYPE: `DataFrame`
`file_path_or_stream`	The path of the internal data table file to be written or a writable file object. TYPE: `str or file object`

khiops.sklearn.dataset

Classes

Dataset(X, y=None, categorical_target=True)

Attributes

table_type property

is_multitable property

Methods:

to_spec()

copy()

get_table(table_name)

create_khiops_dictionary_domain()

create_table_files_for_khiops(output_dir, sort=True)

DatasetTable(name, key=None)

Methods:

check_key()

create_table_file_for_khiops(output_dir, sort=True) abstractmethod

n_features()

create_khiops_dictionary()

PandasTable(name, dataframe, key=None)

NumpyTable(name, array, key=None)

SparseTable(name, matrix, key=None)

Methods:

create_khiops_dictionary()

Functions:

check_dataset_spec(ds_spec)

table_name_of_path(table_path)

get_khiops_type(numpy_type, categorical_str_max_size=None)

get_khiops_variable_name(column_id)

read_internal_data_table(file_path_or_stream, column_dtypes=None)

write_internal_data_table(dataframe, file_path_or_stream)

`Dataset(X, y=None, categorical_target=True)`

`table_type` `property`

`is_multitable` `property`

`to_spec()`

`copy()`

`get_table(table_name)`

`create_khiops_dictionary_domain()`

`create_table_files_for_khiops(output_dir, sort=True)`

`DatasetTable(name, key=None)`

`check_key()`

`create_table_file_for_khiops(output_dir, sort=True)` `abstractmethod`

`n_features()`

`create_khiops_dictionary()`

`PandasTable(name, dataframe, key=None)`

`NumpyTable(name, array, key=None)`

`SparseTable(name, matrix, key=None)`

`create_khiops_dictionary()`

`check_dataset_spec(ds_spec)`

`table_name_of_path(table_path)`

`get_khiops_type(numpy_type, categorical_str_max_size=None)`

`get_khiops_variable_name(column_id)`

`read_internal_data_table(file_path_or_stream, column_dtypes=None)`

`write_internal_data_table(dataframe, file_path_or_stream)`