Hi, thanks for the great paper and open sourced code. It seems a great leap in graph learning. I have one question: my graph is very dynamic, and it won't be feasible to train an embedding for each node in the graph. Can pyHGT generate embedding for previously unseen nodes as graphSage does? Thanks for a clarification.
Hi, thanks for the great paper and open sourced code. It seems a great leap in graph learning. I have one question: my graph is very dynamic, and it won't be feasible to train an embedding for each node in the graph. Can pyHGT generate embedding for previously unseen nodes as graphSage does? Thanks for a clarification.