RGCN and RelationalGraphConvolution

Hi,

I’m following the example from Here. And I’m trying to understand how the RGCN works.

But there is also RGCN Layer

What is the difference between the two implementations, and more important why are they so similar in their names?

It seems that the two links you posted are the same.

1 Like

I meant this RGCN layer from the nn module

and this From the rgcn-hetero implementation

Edit:
There is also this impelementation of an RGCN

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.