PyTorch Geometric is a great library and people should definitely give it a go for themselves. Both libraries implement some of the same algorithms. One of the main differences is that StellarGraph is Tensorflow-based and PyTorch Geometric is, obviously, PyTorch-based. Also, the selection of algorithms is not exactly the same. We are carefully selecting algorithms that achieve state-of-the-art results on common benchmark datasets but also we aim for variety.

StellarGraph is a Python 3 library. The StellarGraph library implements several state-of-the-art algorithms for applying machine learning methods to discover patterns and answer questions using graph-structured data. Added GraphConvolution layer, GCN class for a stack of GraphConvolution layers, and FullBatchNodeGenerator class for feeding data into GCN (from version 0.5.0) We provide examples of using StellarGraph to solve such tasks using several real-world datasets.