Relational data

RelationalData

__init__

__init__(data: Data, schema: Schema) -> None

Relational data structure.

Parameters:

Name	Type	Description	Default
data	Data	A dictionary with table names as keys and pandas.DataFrame's as values.	required
schema	Schema	A Schema object.	required

split

split(
    ratio: float | int | dict[str, float | int],
    reset_index: bool = False,
    rng: NpRng = None,
) -> tuple[RelationalData, RelationalData]

Split the input data according to the given ratios for each root table.

Parameters:

Name	Type	Description	Default
ratio	float | int | dict[str, float | int]	Split ratio. If a float, it must be between 0 and 1. If an integer > 1, then it is interpreted as the number of rows. If a dictionary, it must contain a split ratio for each root table (except LUTs).	required
reset_index	bool	Whether to reset the index of the resulting dataframes.	False
rng	NpRng	Random state. If an int, it will be used as seed, if None the seed will be chosen randomly.	None

Returns:

Type	Description
tuple[RelationalData, RelationalData]	Tuple with the two splits.

to_csv

to_csv(output_dir: Path | str, **kwargs: Any) -> None

Save the RelationalData tables into a directory as csv files.

Parameters:

Name	Type	Description	Default
output_dir	Path | str	Directory where to save the data.	required
**kwargs	Any	Keyword arguments to be passed to pd.to_csv.	{}