Why is the <train_prefix>-class_map.json needed for unsupervised training?
Hi,
I wanted to use GraphSage for unsupervised training but it seems to need the class membership information (<train_prefix>-class_map.json) although I want to run it in unsupervised mode. Did I overlook something or is this class information always needed?
Thanks in advance!
My understanding is that the class_map is needed but it is not used for unsupervised training. You can assign random classes to the nodes.
Yes.. this is just an oversight in the way we set things up. It is not really needed, and we hope to remove the unnecessary dependency soon.
I had the same question. For me, I have two use cases, one is for unsupervised training, and the other for edge classification. I didn't have classes for both. It would be really helpful if you give this information in the README file. Thanks!