Class StreamFeatureView

sdk/python/feast/stream_feature_view.py:46–488

A stream feature view defines a logical group of features that has both a stream data source and a batch data source. Attributes: name: The unique name of the stream feature view. mode: The transformation mode to use for the stream feature view. This can be one of Trans

@typechecked
class StreamFeatureView(FeatureView):
    """
    A stream feature view defines a logical group of features that has both a stream data source and
    a batch data source.

    Attributes:
        name: The unique name of the stream feature view.
        mode: The transformation mode to use for the stream feature view. This can be one of TransformationMode.
        entities: List of entities or entity join keys.
        ttl: The amount of time this group of features lives. A ttl of 0 indicates that
            this group of features lives forever. Note that large ttl's or a ttl of 0
            can result in extremely computationally intensive queries.
        schema: The schema of the feature view, including feature, timestamp, and entity
            columns. If not specified, can be inferred from the underlying data source.
        source: The stream source of data where this group of features is stored.
        aggregations: List of aggregations registered with the stream feature view.
        timestamp_field: Must be specified if aggregations are specified. Defines the timestamp column on which to aggregate windows.
        enable_tiling: Enable tiling optimization for efficient windowed aggregations.
        tiling_hop_size: Time interval between tiles (e.g., 5 minutes). Defaults to 5 minutes.
        online: A boolean indicating whether online retrieval, and write to online store is enabled for this feature view.
        offline: A boolean indicating whether offline retrieval, and write to offline store is enabled for this feature view.
        description: A human-readable description.
        tags: A dictionary of key-value pairs to store arbitrary metadata.
        owner: The owner of the stream feature view, typically the email of the primary maintainer.
        org: The organizational unit that owns this stream feature view (e.g. "ads", "search").
            Defaults to empty string.
        udf: The user defined transformation function. This transformation function should have all of the corresponding imports imported within the function.
        udf_string: The string representation of the user defined transformation function.
        feature_transformation: The transformation to apply to the features.
                Note, feature_transformation has precedence over udf and udf_string.
        stream_engine: Optional dictionary containing stream engine specific configurations.
                Note, it will override the repo-level default stream engine config defined in the yaml file.
    """

    name: str
    entities: List[str]
    ttl: Optional[timedelta]
    source: DataSource
    sink_source: Optional[DataSource] = None
    schema: List[Field]
    entity_columns: List[Field]
    features: List[Field]
    online: bool
    offline: bool
    description: str
    tags: Dict[str, str]
    owner: str
    org: str
    mode: Union[TransformationMode, str]
    materialization_intervals: List[Tuple[datetime, datetime]]
    udf: Optional[FunctionType]
    udf_string: Optional[str]
    feature_transformation: Optional[Transformation]
    stream_engine: Optional[Dict[str, Any]] = None
    aggregations: List[Aggregation]
    timestamp_field: str
    enable_tiling: bool
    tiling_hop_size: Optional[timedelta]