This repository contains the author's implementation in Tensorflow for the paper "Multiview Graph Convolutional Networks with Attention Mechanism".
The structures of MAGCN
The overall structure of MAGCN.
The visualization results
t-SNE visualization for the computed feature representations of a pre-trained model's first hidden layer on the Cora dataset: GCN (left) and our MAGCN (right). Node colors denote classes.
-
Python (>=3.5)
-
Tensorflow (>=1.12.0)
-
Keras (>=2.0.9)
Here we provide the implementation of a MAGCN layer in TensorFlow, along with a minimal execution example (on the Cora dataset). The repository is organised as follows:
data/
contains the necessary dataset files for Cora;models.py
contains the implementation of theMAGCN(Model)
;layers.py
contains the implementation of theMultiGraphConvolution(Layer)
;
Finally, train.py
puts all of the above together and may be used to execute a full training run on Cora.