Supplementary material for the paper " A Graph Similarity for Deep Learning " A Datasets

Here, we describe all the datasets used in our paper with sources. The basic statistics is in Table 1. The benchmarking framework (Dwivedi et al., 2020) provided two community detection datasets, SBM_P A TTERN and SBM_CLUSTER for node classification. The following description is re-phrasing of the explanation in Dwivedi et al. (2020). For SBM_CLUSTER the task is to identify which community each vertex belongs to.