Heimdall.fg.TorchTensorFg

Contents

Heimdall.fg.TorchTensorFg#

class Heimdall.fg.TorchTensorFg(data, embedding_parameters, embedding_filepath=None, **fg_kwargs)[source]#

Bases: PretrainedFg

Mapping of gene names to pretrained embeddings stored as PyTorch tensors.

Attributes

Methods

load_embeddings()

Load the embeddings from disk and process into map.
Parameters:

  • data (CellRepresentation)

  • embedding_parameters (OmegaConf)

  • embedding_filepath (str | PathLike | None)

Contents