DGCNN implementation

[EryiXie]

Hi, I compared your implementation with the original DGCNN and I find out that, in yours the function "get_graph_feature" is only used before the first convolution operation, instead of using it before every convolution layer.

I am wondering if there is some reason behind this difference.

https://github.com/fukexue/RGM/blob/b0e2f742ab58e1b87cabb779d1c24f1441f827c9/models/dgcnn.py#L154-L179