Graph Variational Autoencoder with Pytorch Geometric for Molecule Generation.

This model is based on this paper: https://jcheminf.biomedcentral.com/track/pdf/10.1186/s13321-019-0396-x.pdf

Note that the model is not fully functioning yet.

Please have a look at the original implementation in tensorflow for further directions: https://github.com/seokhokang/graphvae_approx/