simba.read_embedding

simba.read_embedding(path_emb=None, path_entity=None, convert_alias=True, path_entity_alias=None, prefix=None, num_epochs=None)

Read in entity embeddings from pbg training

Parameters:

• path_emb (str, optional (default: None)) – Path to directory for pbg embedding model If None, .settings.pbg_params[‘checkpoint_path’] will be used.

• path_entity (str, optional (default: None)) – Path to entity name file

• prefix (list, optional (default: None)) – A list of entity type prefixes to include. By default, it reads in the embeddings of all entities.

• convert_alias (bool, optional (default: True)) – If True, it will convert entity aliases to the original indices

• path_entity_alias (str, optional (default: None)) – Path to entity alias file

• num_epochs (int, optional (default: None)) – The embedding result associated with num_epochs to read in

Returns:

dict_adata – A dictionary of anndata objects of shape (#entities x #dimensions)

Return type:

dict