Heimdall.fg.Fg

Contents

Heimdall.fg.Fg#

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

Bases: ABC

Abstraction of the gene embedding mapping paradigm.

Parameters:

  • d_embedding (int) – dimensionality of embedding for each gene entity

  • data (CellRepresentation)

  • embedding_parameters (DictConfig)

  • vocab_size (int)

  • pad_value (int)

  • mask_value (int)

  • frozen (bool)

  • rng (int | Generator)

  • do_pca_reduction (bool)

Attributes

adata

identity_valid_mask

Methods

load_from_cache(identity_embedding_index, ...)

Load processed values from cache.

prepare_embedding_parameters()

Replace config placeholders with values after preprocessing.

preprocess_embeddings([float_dtype])

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