BaseFeatureSelectionTransform

class BaseFeatureSelectionTransform(features_to_use: Union[List[str], Literal['all']] = 'all', return_features: bool = False)[source]

Bases: etna.transforms.base.Transform

Base class for feature selection transforms.

Inherited-members

Parameters

  • features_to_use (Union[List[str], Literal['all']]) –

  • return_features (bool) –

Methods

fit(df)

Fit feature model.

fit_transform(df)

May be reimplemented.

inverse_transform(df)

Apply inverse transform to the data.

load(path)

Load an object.

save(path)

Save the object.

to_dict()

Collect all information about etna object in dict.

transform(df)

Select top_k features.
inverse_transform(df: pandas.core.frame.DataFrame) → pandas.core.frame.DataFrame[source]

Apply inverse transform to the data.

Parameters

df (pandas.core.frame.DataFrame) – dataframe to apply inverse transformation

Returns

result – dataframe before transformation

Return type

pd.DataFrame

transform(df: pandas.core.frame.DataFrame) → pandas.core.frame.DataFrame[source]

Select top_k features.

Parameters

df (pandas.core.frame.DataFrame) – dataframe with all segments data

Returns

result – Dataframe with with only selected features

Return type

pd.DataFrame