napistu_torch.labels.labeling_manager

Labeling manager configuration and validation for napistu-torch.

Classes

LabelingManager(*, label_attribute, ...)

Configuration for label-specific featurization strategies.

class napistu_torch.labels.labeling_manager.LabelingManager(*, label_attribute: str, exclude_vertex_attributes: List[str] = <factory>, augment_summary_types: List[str] = <factory>, label_names: Dict[int, ~typing.Any] | None=None)

Bases: BaseModel

Configuration for label-specific featurization strategies.

This class organizes and validates the attributes needed for different labeling approaches in molecular network analysis tasks.

label_attribute

The vertex attribute to use as the target label

Type:

str

exclude_vertex_attributes

Vertex attributes to exclude from feature extraction

Type:

List[str]

augment_summary_types

SBML DFS summary types to add during graph augmentation. Used by augment_napistu_graph when calling add_sbml_dfs_summaries.

Type:

List[str]

label_names

Optional lookup table mapping label integers to their original names. Used to track the mapping between encoded integers and original label values.

Type:

Optional[Dict[int, Any]]

Public Methods

--------------

get_label_names

Get the label names mapping, returning empty dict if None

to_dict

Convert the labeling strategy to a dictionary

from_dict(config)

Create a LabelingManager from a dictionary configuration

classmethod from_dict(config: Dict[str, Any]) → LabelingManager

Create a LabelingManager from a dictionary configuration.

classmethod validate_exclude_attributes(v: List[str]) → List[str]

Validate that excluded attributes are strings.

classmethod validate_label_attribute(v: str) → str

Validate that the label attribute is not empty.

classmethod validate_summary_types(v: List[str]) → List[str]

Validate that summary types are valid SBML DFS summary types.

get_label_names() → Dict[int, Any]

Get the label names mapping, returning empty dict if None.

to_dict() → Dict[str, Any]

Convert the labeling strategy to a dictionary.

_abc_impl = <_abc._abc_data object>

augment_summary_types: List[str]

exclude_vertex_attributes: List[str]

label_attribute: str

label_names: Dict[int, Any] | None

model_config = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].