Heimdall.fg.IdentityFg

Contents

Heimdall.fg.IdentityFg#

class Heimdall.fg.IdentityFg(data, embedding_parameters, d_embedding, vocab_size, pad_value=None, mask_value=None, frozen=False, rng=0, do_pca_reduction=True)[source]#

Bases: Fg

Identity mapping of gene names to embeddings.

This is the simplest possible Fg; it implies the use of learnable gene embeddings that are initialized randomly, as opposed to the use of pretrained embeddings.

Attributes

Methods

preprocess_embeddings([float_dtype])

Preprocess gene embeddings and store them for use during model inference.
Parameters:

  • data (CellRepresentation)

  • embedding_parameters (DictConfig)

  • d_embedding (int)

  • vocab_size (int)

  • pad_value (int)

  • mask_value (int)

  • frozen (bool)

  • rng (int | Generator)

  • do_pca_reduction (bool)

Contents