class src.gconv.Siamese_Gconv(in_features, num_features)[source]

Siamese Gconv neural network for processing arbitrary number of graphs.

  • in_features – the dimension of input node features

  • num_features – the dimension of output node features

forward(g1: Tuple[torch.Tensor, torch.Tensor, torch.Tensor, int], *args) Union[torch.Tensor, List[torch.Tensor]][source]

Forward computation of Siamese Gconv.

  • g1 – The first graph, which is a tuple of (\((b\times n\times n)\) {0,1} adjacency matrix, \((b\times n\times d)\) input node embedding, normalize connectivity matrix or not)

  • args – Other graphs


A list of tensors composed of new node embeddings \((b\times n\times d^\prime)\)

training: bool