MCPcopy Create free account
hub / github.com/XiaShan1227/Graphormer / forward

Method forward

layer.py:239–259  ·  view source on GitHub ↗

h′(l) = MHA(LN(h(l−1))) + h(l−1) h(l) = FFN(LN(h′(l))) + h′(l) :param x: node feature matrix :param edge_attr: edge feature matrix :param b: spatial Encoding matrix :param edge_paths: pairwise node paths in edge indexes :param ptr: batch poin

(self,
                x: torch.Tensor,
                edge_attr: torch.Tensor,
                b: torch,
                edge_paths,
                ptr)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected