logml.data.transformers.utils

Functions

`get_affected_categorical_columns`(self, dataframe)	Finds all categorical columns.
`get_affected_numerical_columns`(self, dataframe)	Finds all numerical columns.

Classes

`EligibleTransformers`(args, *kwargs)	Provides registry functionality for Transformers (feature extraction utilities).
`Obsolete_NN`()	This transformer is osbolete.

Exceptions

TransformerException

Exception raised when there is an Exception caught while using TransformerAPI.

exception logml.data.transformers.utils.TransformerException

Bases: Exception

Exception raised when there is an Exception caught while using TransformerAPI.

class logml.data.transformers.utils.EligibleTransformers(*args, **kwargs)

Bases: logml.common.BaseEligibleRegistrator

Provides registry functionality for Transformers (feature extraction utilities).

get_transformer(name, params: dict = None, metadata_cfg: logml.configuration.modeling.ModelingTaskSpec = None, cfg=None, global_params: dict = None, logger=None) → BaseTransformer

Instantiate a transformer.

Outdated transformers and their params bay be mapped to actual ones.

map_request(name, params) → Tuple[str, dict]: Map old transformer to the new one. Backward compatibility for config

mapping: Dict[str, Type]

class logml.data.transformers.utils.Obsolete_NN

Bases: object

This transformer is osbolete.

logml.data.transformers.utils.get_affected_categorical_columns(self: BaseTransformer, dataframe: pandas.core.frame.DataFrame, dataset_metadata: logml.data.metadata.DatasetMetadata = None) → List[str]

Finds all categorical columns.

WARNING: deprecated and will be replaced with category filtering. Uses self. reference.

logml.data.transformers.utils.get_affected_numerical_columns(self, dataframe: pandas.core.frame.DataFrame, dataset_metadata: Optional[logml.data.metadata.DatasetMetadata] = None) → List[str]

Finds all numerical columns.

WARNING: deprecated and will be replaced with category filtering. uses self. reference.