feature_encoders.validate package

Submodules

feature_encoders.validate.schemas module

class feature_encoders.validate.schemas.CategoricalSchema(*, type: str, feature: str, max_n_categories: int = None, stratify_by: Optional[Union[str, List[str]]] = None, excluded_categories: Optional[Union[str, List[str]]] = None, unknown_value: int = None, min_samples_leaf: int = 1, max_features: Union[str, int, float] = 'auto', random_state: int = None, encode_as: str = 'onehot')[source]

Bases: pydantic.main.BaseModel

classmethod check_encode_as(data)[source]

classmethod check_lists(data)[source]

classmethod check_max_features(data)[source]

encode_as: str

excluded_categories: Optional[Union[str, List[str]]]

feature: str

max_features: Union[str, int, float]

max_n_categories: Optional[int]

min_samples_leaf: int

random_state: Optional[int]

stratify_by: Optional[Union[str, List[str]]]

type: str

unknown_value: Optional[int]

class feature_encoders.validate.schemas.CyclicalSchema(*, type: str, seasonality: str, ds: str = None, period: float = None, fourier_order: int = None, remainder: str = 'passthrough', replace: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

ds: Optional[str]

fourier_order: Optional[int]

period: Optional[float]

remainder: str

replace: bool

seasonality: str

type: str

class feature_encoders.validate.schemas.DatetimeSchema(*, type: str, ds: str = None, remainder: str = 'passthrough', replace: bool = False, subset: Optional[Union[str, List[str]]] = None)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

classmethod check_subset(data)[source]

ds: Optional[str]

remainder: str

replace: bool

subset: Optional[Union[str, List[str]]]

type: str

class feature_encoders.validate.schemas.LinearSchema(*, type: str, feature: str, as_filter: bool = False, include_bias: bool = False)[source]

Bases: pydantic.main.BaseModel

as_filter: bool

feature: str

include_bias: bool

type: str

class feature_encoders.validate.schemas.SplineSchema(*, type: str, feature: str, n_knots: int = 5, degree: int = 3, strategy: Optional[Union[str, List]] = 'uniform', extrapolation: str = 'constant', include_bias: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_extrapolation(data)[source]

classmethod check_strategy(data)[source]

degree: Optional[int]

extrapolation: Optional[str]

feature: str

include_bias: bool

n_knots: Optional[int]

strategy: Optional[Union[str, List]]

type: str

class feature_encoders.validate.schemas.TrendSchema(*, type: str, ds: str = None, name: str = 'growth', remainder: str = 'passthrough', replace: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

ds: Optional[str]

name: str

remainder: str

replace: bool

type: str

Module contents

class feature_encoders.validate.CategoricalSchema(*, type: str, feature: str, max_n_categories: int = None, stratify_by: Optional[Union[str, List[str]]] = None, excluded_categories: Optional[Union[str, List[str]]] = None, unknown_value: int = None, min_samples_leaf: int = 1, max_features: Union[str, int, float] = 'auto', random_state: int = None, encode_as: str = 'onehot')[source]

Bases: pydantic.main.BaseModel

classmethod check_encode_as(data)[source]

classmethod check_lists(data)[source]

classmethod check_max_features(data)[source]

encode_as: str

excluded_categories: Optional[Union[str, List[str]]]

feature: str

max_features: Union[str, int, float]

max_n_categories: Optional[int]

min_samples_leaf: int

random_state: Optional[int]

stratify_by: Optional[Union[str, List[str]]]

type: str

unknown_value: Optional[int]

class feature_encoders.validate.CyclicalSchema(*, type: str, seasonality: str, ds: str = None, period: float = None, fourier_order: int = None, remainder: str = 'passthrough', replace: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

ds: Optional[str]

fourier_order: Optional[int]

period: Optional[float]

remainder: str

replace: bool

seasonality: str

type: str

class feature_encoders.validate.DatetimeSchema(*, type: str, ds: str = None, remainder: str = 'passthrough', replace: bool = False, subset: Optional[Union[str, List[str]]] = None)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

classmethod check_subset(data)[source]

ds: Optional[str]

remainder: str

replace: bool

subset: Optional[Union[str, List[str]]]

type: str

class feature_encoders.validate.LinearSchema(*, type: str, feature: str, as_filter: bool = False, include_bias: bool = False)[source]

Bases: pydantic.main.BaseModel

as_filter: bool

feature: str

include_bias: bool

type: str

class feature_encoders.validate.SplineSchema(*, type: str, feature: str, n_knots: int = 5, degree: int = 3, strategy: Optional[Union[str, List]] = 'uniform', extrapolation: str = 'constant', include_bias: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_extrapolation(data)[source]

classmethod check_strategy(data)[source]

degree: Optional[int]

extrapolation: Optional[str]

feature: str

include_bias: bool

n_knots: Optional[int]

strategy: Optional[Union[str, List]]

type: str

class feature_encoders.validate.TrendSchema(*, type: str, ds: str = None, name: str = 'growth', remainder: str = 'passthrough', replace: bool = False)[source]

Bases: pydantic.main.BaseModel

classmethod check_remainder(data)[source]

ds: Optional[str]

name: str

remainder: str

replace: bool

type: str