Containers#
Containers are the data structures returned by the extractors and used by the iterators.
They are udao’s vehicle for data along the processing pipeline.
Base Container#
- class udao.data.containers.base_container.BaseContainer#
Bases:
ABCBase class for containers. Containers are used to store and retrieve data from a dataset, based on a key.
Query Structure Container#
- class udao.data.containers.query_structure_container.QueryDescription(template_id: int, template_graph: dgl.heterograph.DGLGraph, graph_features: Iterable, meta_features: Iterable | None, operation_types: Iterable)#
Bases:
object
- class udao.data.containers.query_structure_container.QueryStructureContainer(graph_features: DataFrame, graph_meta_features: DataFrame | None, template_plans: Dict[int, QueryPlanStructure], key_to_template: Dict[str, int], operation_types: Series)#
Bases:
BaseContainerContainer for the query structure and features of a query plan.
- graph_features: DataFrame#
Stores the features of the operations in the query plan.
- graph_meta_features: DataFrame | None#
Stores the meta features of the operations in the query plan.
- key_to_template: Dict[str, int]#
Link a key to a template id.
- operation_types: Series#
Stores the operation types of the operations in the query plan.
- template_plans: Dict[int, QueryPlanStructure]#
Link a template id to a QueryPlanStructure
Dataframe Container#
- class udao.data.containers.tabular_container.TabularContainer(data: DataFrame)#
Bases:
BaseContainerContainer for tabular data, stored in DataFrame format.