Error in Keras_FGSD

[Anwar-Said]

Hi,
Computing graph Laplacian using np.diag(D)-A is incorrect. Please make it correct.

Anwar,
We are currently working on the efficient implementation of FGSD. Our implementation will resolve this issue as well as will provide the sparse representations that will be scalable on significantly large graphs.