pipeline.Embedding.models package#

A package that defines various embedding networks.

pipeline.Embedding.models.get_model(model_type=None)[source]#

Get an embedding model class from its name.

Parameters:

model_type (Optional[str]) – embedding type. Available so far are only layerless.

Return type:

Type[EmbeddingBase]

Returns:

The GNN model class that can be instantiated.

pipeline.Embedding.models.layerless_embedding module#

class pipeline.Embedding.models.layerless_embedding.LayerlessEmbedding(hparams)[source]#

Bases: EmbeddingBase

forward(x)[source]#
property n_hiddens: int#

Number of hidden units for each hidden layer.

Correspond to the hyperparameter n_hiddens. Falls back to emb_hidden otherwise.

property n_layers: int#

Number of layers.

Correspond to the hyperparameter n_layers. Falls back to nb_layer otherwise.