Scalability for large graphs

tkipf / gcn

Implementation of Graph Convolutional Networks in TensorFlow

MIT License

7.1k stars 2k forks source link

Scalability for large graphs #190

Open jeskreiswinkler-etsy opened 3 years ago

jeskreiswinkler-etsy commented 3 years ago

I am interested in using this code for analyzing large graphs, but it seems like memory complexity is going to scale like a polynomial if this adjacency matrix code is used:

adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))

Is there an alternative?