Hi, I have an heterogeneous graph (two node types and three edge types).
In the function multi_update_all I supply the etype_dict (that is, a pair of (message_func, reduce_func) per etype) and use ‘stack’ as cross_reducer. Afterwards I would like to apply the node_func to each ntype (processing the stacked representation obtained in each node per relation), however node_func is called only for one of the two node types.
Being ‘multi_update_all’ thought for an heterogeneous graph, shouldn’t node_func be called for every node type?
Thanks!
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.