GNN CheatSheet

[nkamalesh111]

Hello everyone, I am new to this field. I am going through the documentation and there I found this : GNN Cheatsheet: https://pytorch-geometric.readthedocs.io/en/latest/cheatsheet/gnn_cheatsheet.html. Please explain what is the purpose of this as it shows a table and there are tick marks. Please explain how to understand the first function with the related attributes and then I can proceed further.Thank you. Regards Kamalesh.

[wsad1]

Each tick indicates that the GNN supports that features. For example take GCNConv,

1. it supports passing message passing based on SparseTensor, which is a efficient format to represent edges
2. it supports having weight on the edges so edge_weight is ticked
3. but its doesn't support edge attributes i.e multi-deimensional features on the edge so edge_attr is not ticked,
4. bipartite graphs are not supported as input. Read more about such graphs here
5. lazy is ticked, as you don't need to define the number of input features while initializing GCNConv it will autoinfer the number of inputs when you first call forward.