get_numeric_roles_stat

sparklightautoml.reader.guess_roles.get_numeric_roles_stat(train, subsample=100000, random_state=42, manual_roles=None)

Calculate statistics about different encodings performances.

We need it to calculate rules about advanced roles guessing. Only for numeric data.

Parameters:

• train (SparkDataset) – Dataset.

• subsample (Union[float, int, None]) – size of subsample.

• random_state (int) – int.

• manual_roles (Optional[Dict[str, ColumnRole]]) – Dict.

Return type:

DataFrame

Returns:

DataFrame.