High Skip Network Implementation model for binary node classification.
- class topomodelx.nn.simplicial.hsn.HSN(channels, n_layers=2)[source]#
High Skip Network Implementation for binary node classification.
- Parameters:
- channelsint
Dimension of features.
- n_layersint
Amount of message passing layers.
- forward(x_0, incidence_1, adjacency_0)[source]#
Forward computation.
- Parameters:
- x_0torch.Tensor, shape = (n_nodes, channels)
Node features.
- incidence_1torch.Tensor, shape = (n_nodes, n_edges)
Boundary matrix of rank 1.
- adjacency_0torch.Tensor, shape = (n_nodes, n_nodes)
Adjacency matrix (up) of rank 0.
- Returns:
- torch.Tensor, shape = (n_nodes, channels)
Final node hidden representations.