1 data_handling.ase_efm.semantics

data_handling.ase_efm.semantics

Normalize EFM semantic-id metadata for data and visualization consumers.

This module provides tolerant normalization of sparse mappings or legacy dense name sequences and a single class-name lookup with a numeric fallback. It owns presentation-facing canonicalization only; it does not remap stored semantic ids, infer categories, or define the dataset ontology.

Sparse mappings and legacy dense sequences are converted to one integer-keyed representation without treating unknown numeric ids as meaningful labels.

1.1 Attributes

Name Description
SemanticNameMap Sparse mapping from EFM semantic id to human-readable class name.

1.2 Functions

Name Description
normalize_semantic_name_map Normalize raw EFM semantic-name metadata into a sparse id map.
semantic_class_name Return a display class name for one EFM semantic id.

1.2.1 normalize_semantic_name_map

data_handling.ase_efm.semantics.normalize_semantic_name_map(value)

Normalize raw EFM semantic-name metadata into a sparse id map.

1.2.1.1 Parameters

Name Type Description Default
value Mapping[object, object] | Sequence[object] | None Raw semantic metadata from EFM, msgpack, or legacy dense lists. required

1.2.1.2 Returns

Name Type Description
SemanticNameMap | None Sparse {semantic_id: class_name} mapping, or None when missing.

1.2.2 semantic_class_name

data_handling.ase_efm.semantics.semantic_class_name(sem_id, sem_id_to_name)

Return a display class name for one EFM semantic id.

Back to top