Skip to main content

hugging_face_ner

Hugging Face Named Entity Recognition (NER) Algorithm.

Classes​

HuggingFaceNERInference​

class HuggingFaceNERInference(    datastructure: DataStructure,    model_id: str,    batch_size: int = 16,    aggregation_strategy: _AggregationStrategy = 'simple',    seed: int = 42,    access_token: Optional[str] = None,    postprocessors: Optional[list[dict[str, Any]]] = None,):

Inference for pre-trained Hugging Face Named Entity Recognition (NER) models.

This algorithm extracts named entities from text using HuggingFace's token-classification pipeline.

Arguments

  • ****kwargs**: Additional keyword arguments.
  • aggregation_strategy: Strategy to aggregate tokens into entities. Options are: "simple", "first", "average", "max", "none". Defaults to "simple".
  • batch_size: The batch size for inference. Defaults to 16.
  • datastructure: The data structure to use for the algorithm.
  • model_id: HuggingFace model identifier for the NER model.
  • seed: Sets the seed of the algorithm for reproducibility. Defaults to 42.

Attributes

  • aggregation_strategy: The token aggregation strategy.
  • batch_size: The batch size for inference.
  • class_name: The name of the algorithm class.
  • fields_dict: A dictionary mapping all attributes that will be serialized in the class to their marshmallow field type. (e.g. fields_dict = {"class_name": fields.Str()}).
  • model_id: The HuggingFace model identifier being used.
  • nested_fields: A dictionary mapping all nested attributes to a registry that contains class names mapped to the respective classes. (e.g. nested_fields = {"datastructure": datastructure.registry})
  • seed: The random seed for reproducibility.

Ancestors​

Variables​

Methods​

create​

def create(self, role: Union[str, Role], **kwargs: Any) ‑> Any:

Create an instance representing the role specified.

modeller​

def modeller(    self, *, context: ProtocolContext, **kwargs: Any,) ‑> bitfount.federated.algorithms.hugging_face_algorithms.base._HFModellerSide:

Returns the modeller side of the HuggingFaceNERInference algorithm.

worker​

def worker(    self, *, context: ProtocolContext, **kwargs: Any,) ‑> bitfount.federated.algorithms.hugging_face_algorithms.hugging_face_ner._WorkerSide:

Returns the worker side of the HuggingFaceNERInference algorithm.