class EmbeddingFactory

Bases: object

Producing unsupervised embeddings for a given dataset

build_embeddings(loss_name: str, conv: str, data: List[Graph], device: torch.device, number_of_trials: int, tune_out: bool = False) → numpy.typing.NDArray

Build embeddings based on passed dataset and settings

Parameters:

  • loss_name – (str): Name of loss function for embedding learning in GeomGCN layer

  • conv – (str) Name of convolution used in unsupervied embeddings

  • data – (Graph): Input Graph

  • device – (device): Device ‘cuda’ or ‘cpu’

  • (int) (number_of_trials) – Number of trials for optuna tuning embeddings

  • (bool) (tune_out) – Flag if you want tune out layer of embeddings

Returns:

(NDArray) embeddings NumPy array of (N_nodes) x (N_emb_dim)