Hierarchical Graph Representation Learning with Differentiable Pooling
Diffpool implementation for the Graph Deep Learning for Medical Applications seminar in the summer semester of 2022 at the Technical University of Munich.
Links to the original paper:
- ArXiv: https://arxiv.org/abs/1806.08804
- Code Repository: https://github.com/RexYing/diffpool
Official Pytorch Geometric (PyG) example: https://github.com/pyg-team/pytorch_geometric/blob/master/examples/proteins_diff_pool.py
Official Deep Graph Library (DGL) example: https://github.com/dmlc/dgl/tree/master/examples/pytorch/diffpool
Dependencies
- PyTorch 1.11+
- PyTorch-Geometric 2.0+
- Deep Graph Library 0.8+
Installation
Make sure you have Python 3.9 and Poetry installed on your system. The code is only tested on Linux (Fedora) and Windows 11.
To install all dependencies, run:
poetry install