scSLAT.model.graphconv.combnet.CombUnweighted

class scSLAT.model.graphconv.combnet.CombUnweighted(K=1, cached=False, bias=True, **kwargs)[source]

Bases: MessagePassing

LGCN (GCN without learnable and concat)

Parameters:

K (typing.Optional[int]) – K-hop neighbor to propagate

Methods

forward

Runs the forward pass of the module.

message

Constructs messages from node \(j\) to node \(i\) in analogy to \(\phi_{\mathbf{\Theta}}\) for each edge in edge_index.

Attributes