You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
My implementation of the original GAT paper (Veličković et al.). I've additionally included the playground.py file for visualizing the Cora dataset, GAT embeddings, an attention mechanism, and entropy histograms. I've supported both Cora (transductive) and PPI (inductive) examples!
[ICLR 2020; IPDPS 2019] Fast and accurate minibatch training for deep GNNs and large graphs (GraphSAINT: Graph Sampling Based Inductive Learning Method).
This is a PyTorch implementation of the GeniePath model in <GeniePath: Graph Neural Networks with Adaptive Receptive Paths> (https://arxiv.org/abs/1802.00910)