Hierarchical Inter-Message Passing for Learning on Molecular Graphs

---

This is a PyTorch implementation of Hierarchical Inter-Message Passing for Learning on Molecular Graphs, as described in our paper:

Matthias Fey, Jan-Gin Yuen, Frank Weichert: Hierarchical Inter-Message Passing for Learning on Molecular Graphs (GRL+ 2020)

Requirements

• PyTorch (>=1.4.0)
• PyTorch Geometric (>=1.5.0)
• OGB (>=1.1.0)

Experiments

Experiments can be run via:

$ python train_zinc_subset.py
$ python train_zinc_full.py
$ python train_hiv.py
$ python train_muv.py
$ python train_tox21.py
$ python train_ogbhiv.py
$ python train_ogbpcba.py

Cite

Please cite our paper if you use this code in your own work:

@inproceedings{Fey/etal/2020,
  title={Hierarchical Inter-Message Passing for Learning on Molecular Graphs},
  author={Fey, M. and Yuen, J. G. and Weichert, F.},
  booktitle={ICML Graph Representation Learning and Beyond (GRL+) Workhop},
  year={2020},
}