Skip to main content

timm_inference

Hugging Face TIMM inference Algorithm.

Adapted from: https://github.com/huggingface/api-inference-community/

Classes​

TIMMInference​

class TIMMInference(    datastructure: DataStructure,    model_id: str,    num_classes: int | None = None,    batch_size: int = 1,    checkpoint_path: os.PathLike[str] | str | None = None,    class_outputs: list[str] | None = None,    postprocessors: list[dict[str, Any]] | None = None,):

HuggingFace TIMM Inference Algorithm.

Arguments

  • **kwargs: Additional keyword arguments.
  • batch_transformations: A list of dictionaries containing the batch transformations. Defaults to None.
  • checkpoint_path: The path to a checkpoint file local to the Pod. Defaults to None.
  • class_outputs: A list of explict class outputs to use as labels. Defaults to None.
  • datastructure: The data structure to use for the algorithm.
  • model_id: The model id to use from the Hugging Face Hub.
  • num_classes: The number of classes in the model. Defaults to None.

Attributes

  • checkpoint_path: The path to a checkpoint file local to the Pod. Defaults to None.
  • class_name: The name of the algorithm class.
  • class_outputs: A list of explict class outputs to use as labels. Defaults to None.
  • 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 model id to use from the Hugging Face Hub.
  • 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})
  • num_classes: The number of classes in the model. Defaults to None.

Ancestors​

Variables​

Methods​

create​

def create(self, role: 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 TIMMInference algorithm.

worker​

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

Returns the worker side of the TIMMInference algorithm.