From Fastai to Graph Neural Nets

Hi everyone. I am sorry if I’m repeating a previous discussion. I wasn’t able to find a corresponding answer to my question below.

I’ve recently started working on graph learning. On a daily basis, I am dealing with lots of graph-related problems and I think I can experiment with graph neural nets to produce some intuitive solutions. In general, I’ll be interested in graph classification and embeddings.

I am aware of some of the frameworks that are built on PyTorch. My question is, would it be wise to start with Fastai (course, book, library) as a beginner and eventually move to graph learning? If you think so, do you have any recommendations for how to transition? And finally, do you think I should start with the 2022 version of the course starting next week?

Thanks in advance.

Which graph database are you using?