giannisnik / cnn-graph-classification

A convolutional neural network for graph classification in PyTorch

Kernel Graph Convolutional Neural Networks

Code for the paper Kernel Graph Convolutional Neural Networks.

Requirements

Code is written in Python 3.6 and requires:

• PyTorch 0.3
• NetworkX 1.11
• igraph 0.7
• scikit-learn 0.18

Datasets

Use the following link to download datasets:

https://ls11-www.cs.tu-dortmund.de/staff/morris/graphkerneldatasets

Extract the datasets into the datasets folder.

Run the model

First, specify the dataset and the hyperparameters in the main.py file. Then, use the following command:

$ python main.py

Cite

Please cite our paper if you use this code:

@inproceedings{nikolentzos2018kernel,
  title={Kernel Graph Convolutional Neural Networks},
  author={Nikolentzos, Giannis and Meladianos, Polykarpos and Tixier, Antoine Jean-Pierre and Skianis, Konstantinos and Vazirgiannis, Michalis},
  booktitle={International Conference on Artificial Neural Networks},
  pages={22--32},
  year={2018},
  organization={Springer}
}

---

Provided for academic use only