Implementation of the forward-forward procedure for GNNs described by Paliotta et. al. in the paper titled "Graph Neural Networks Go Forward-Forward" released on 10 Feb 2023. This framework allows training graph neural networks with forward passes only, without backpropagation.
Curently a work in progress