Official code for "InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization" (ICLR 2020, spotlight)
I don't have any labeled data.y, so how to learn? Is this unsupervised learning?
I'm finding a fully unsupervised graph-level clustering/classification method.
If data.y should be needed to train, how many graph-level data have to have its label?
I have millions of graph data that cannot be labeled by human efforts.
In the page https://github.com/sunfanyunn/InfoGraph/blob/master/unsupervised/gin.py
I don't have any labeled data.y, so how to learn? Is this unsupervised learning? I'm finding a fully unsupervised graph-level clustering/classification method.
If data.y should be needed to train, how many graph-level data have to have its label? I have millions of graph data that cannot be labeled by human efforts.