Skip to content

featurebyte.Feature

class Feature(
*,
user_id: Union[PydanticObjectId, NoneType]=None,
name: Union[StrictStr, NoneType]=None,
created_at: Union[datetime, NoneType]=None,
updated_at: Union[datetime, NoneType]=None,
block_modification_by: List[ReferenceInfo]=None,
description: Union[StrictStr, NoneType]=None,
graph: QueryGraph=None,
node_name: str,
tabular_source: TabularSource,
feature_store: FeatureStoreModel,
catalog_id: PydanticObjectId=None,
dtype: DBVarType
)

Description

A feature is input data that is used to train Machine Learning models and compute predictions. These features can sometimes be derived from attributes already present in the source tables. For instance, a customer churn model can use features obtained directly from a customer profile table, such as age, gender, income, and location.

However, in most cases, features are created by applying a series of row transformations, joins, filters, and aggregates. For instance, a customer churn model may utilize aggregate features that reflect the customer's account details over a given period, such as the entropy of product types purchased over the past 12 weeks, the count of canceled orders over the past 56 weeks, and the amount of money spent over the past 7 days.

Parameters

  • user_id: Union[PydanticObjectId, NoneType]

  • name: Union[StrictStr, NoneType]

  • created_at: Union[datetime, NoneType]

  • updated_at: Union[datetime, NoneType]

  • block_modification_by: List[ReferenceInfo]

  • description: Union[StrictStr, NoneType]

  • graph: QueryGraph

  • node_name: str

  • tabular_source: TabularSource

  • feature_store: FeatureStoreModel

  • catalog_id: PydanticObjectId

  • dtype: DBVarType