search

tkipf / keras-gcn

Keras implementation of Graph Convolutional Networks
MIT License
791 stars 266 forks source link

About the call() in the graph.py #23

Closed zhangxiaozao closed 6 years ago

zhangxiaozao commented 6 years ago

I have several questions about the call() in the graph.py 1>Does the "basis" represent for Aˆ (Aˆ = D˜ ^-0.5 A˜D˜ ^-0.5) ? 2>I saw the "support"=1, does it mean that the "supports" only contains one example's new feature ?If so, where are other examples' new features?

Besides, for the first graph layer, the "inputs" contains Aˆ, but for the second graph layer, how is Aˆ contained in the "inputs"? Thank you very much! Look forward for your reply! Thanks!

tkipf commented 6 years ago

Yes, the variable basis for the simple GCN model represents the renormalized adjacency matrix. Support=1 just means we're using a single basis. Support > 1 is required if you, e.g., use a polynomial basis with multiple terms.