Preprocessors

TableConfig pydantic-config

TableConfig(
    description: str | None = None,
    column_descriptions: dict[str, str] = {},
    config: dict[str, Any] = ConfigDict(),
)

A configuration for a table preprocessor.

Parameters:

Name	Type	Description	Default
description	str | None	A description of the table, used in the corresponding JSON schema.	None
column_descriptions	dict[str, str]	A dictionary with an optional description for the table columns, used in the corresponding JSON schema.	{}
config	dict[str, Any]	A pydantic.ConfigDict to be used in the corresponding Pydantic model.	ConfigDict()

TablePreproc

from_table classmethod

from_table(
    table: Table,
    *,
    config: TableConfig | None = None,
    overwrites: dict[str, ColumnPreproc | None]
    | None = None,
) -> TablePreproc

Create the table preprocessor from a Table object.

Parameters:

Name	Type	Description	Default
table	Table	The Table object with the column information.	required
config	TableConfig | None	The table preprocessor configuration.	None
overwrites	dict[str, ColumnPreproc | None] | None	Optional overwrites for each column. A value of None means the column is ignored.	None

Returns:

Type	Description
TablePreproc	The table preprocessor.

get_default_column classmethod

get_default_column(
    column: Column | str,
) -> type[ColumnPreproc] | None

Get the type of the default column preprocessor for the given Column.

Parameters:

Name	Type	Description	Default
column	Column | str	The column type, as a Column instance.	required

Returns:

Type	Description
type[ColumnPreproc] | None	The type of the column preprocessor. Returns None if the column is not supported.

RelSynthPreproc

from_schema classmethod

from_schema(
    schema: Schema,
    *,
    config: dict[str, TableConfig] | None = None,
    overwrites: dict[str, TablePreproc | None]
    | None = None,
) -> RelSynthPreproc

Build the preprocessor from a Schema object.

Parameters:

Name	Type	Description	Default
schema	Schema	The dataset Schema.	required
config	dict[str, TableConfig] | None	A dictionary with optional table preprocessor configurations.	None
overwrites	dict[str, TablePreproc | None] | None	Optional overwrites for each table. A value of None means the table is ignored.	None

Returns:

Type	Description
RelSynthPreproc	The preprocessor.

from_data classmethod

from_data(
    data: RelationalData,
    *,
    config: dict[str, TableConfig] | None = None,
    overwrites: dict[str, TablePreproc | None]
    | None = None,
) -> RelSynthPreproc

Build and fit the preprocessor from a RelationalData.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data used to build and fit the preprocessor.	required
config	dict[str, TableConfig] | None	A dictionary with optional table preprocessor configurations.	None
overwrites	dict[str, TablePreproc | None] | None	Optional overwrites for each table. A value of None means the table is ignored.	None

Returns:

Type	Description
RelSynthPreproc	The fitted preprocessor.

fit

fit(data: RelationalData) -> RelSynthPreproc

Fit the preprocessor to the input data.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data to fit the preprocessor.	required

Returns:

Type	Description
RelSynthPreproc	The fitted preprocessor.

protect

protect(data: RelationalData, fit: bool) -> RelationalData

Protect the data and optionally fit the protectors.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data to protect.	required
fit	bool	Whether to fit the protectors.	required

Returns:

Type	Description
RelationalData	The protected data.

RelEventPreproc

from_schema classmethod

from_schema(
    schema: Schema,
    *,
    ord_cols: dict[str, str] | None = None,
    config: dict[str, TableConfig] | None = None,
    overwrites: dict[str, TablePreproc | None]
    | None = None,
    **kwargs: Any,
) -> RelEventPreproc

Build the preprocessor from a Schema object.

Parameters:

Name	Type	Description	Default
schema	Schema	The dataset Schema.	required
ord_cols	dict[str, str] | None	For each table, the column that contain the order of the event. If None, only a single event table is possible, and the events are considered ordered as for the order in the event table. If a dictionary, each event table must contain an order column.	None
config	dict[str, TableConfig] | None	A dictionary with optional table preprocessor configurations.	None
overwrites	dict[str, TablePreproc | None] | None	Optional overwrites for each table. A value of None means the table is ignored.	None

Returns:

Type	Description
RelEventPreproc	The preprocessor.

from_data classmethod

from_data(
    data: RelationalData,
    *,
    ord_cols: dict[str, str] | None = None,
    config: dict[str, TableConfig] | None = None,
    overwrites: dict[str, TablePreproc | None]
    | None = None,
    **kwargs: Any,
) -> RelEventPreproc

Build and fit the preprocessor from a RelationalData.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data used to build and fit the preprocessor.	required
ord_cols	dict[str, str] | None	For each table, the column that contain the order of the event. If None, only a single event table is possible, and the events are considered ordered as for the order in the event table. If a dictionary, each event table must contain an order column.	None
config	dict[str, TableConfig] | None	A dictionary with optional table preprocessor configurations.	None
overwrites	dict[str, TablePreproc | None] | None	Optional overwrites for each table. A value of None means the table is ignored.	None

Returns:

Type	Description
RelEventPreproc	The fitted preprocessor.

fit

fit(data: RelationalData) -> RelEventPreproc

Fit the preprocessor to the input data.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data to fit the preprocessor.	required

Returns:

Type	Description
RelSynthPreproc	The fitted preprocessor.

protect

protect(data: RelationalData, fit: bool) -> RelationalData

Protect the data and optionally fit the protectors.

Parameters:

Name	Type	Description	Default
data	RelationalData	The data to protect.	required
fit	bool	Whether to fit the protectors.	required

Returns:

Type	Description
RelationalData	The protected data.