This is an implementation of Learning Deep Generative Models of Graphs by Yujia Li, Oriol Vinyals, Chris Dyer, Razvan Pascanu, Peter Battaglia.
For molecule generation, see DGL-LifeSci.
- Python 3.5.2
- Pytorch 0.4.1
- Matplotlib 2.2.2
python3 main.py
90% accuracy for cycles compared with 84% accuracy reported in the original paper.
On AWS p3.2x instance (w/ V100), one epoch takes ~526s.
We would like to thank Yujia Li for providing details on the implementation.