Adapted from cynricfu/MAGNN.
We add GCN and GAt comparison under DBLP dataset for now. Ohter experiments to be completed.
- torch 1.4.0 cuda 10.1
- dgl 0.4.3 cuda 10.1
- networkx 2.3
- scikit-learn 0.23.2
- scipy 1.5.2
- Download DBLP_processed.zip from tsinghua-cloud or google-drive
- mkdir checkpoint
- mkdir data
- mkdir data/preprocessed
- unzip DBLP_Processed.zip to data/preprocessed
- run run_DBLP_gnn.py
python run_DBLP_gnn.py --model gat
python run_DBLP_gnn.py --model gcnGCN is not good enough for now. Try to modify and tune later.
| micro f1 score | macro f1 score | |
|---|---|---|
| MAGNN | ~93.5 | ~93 |
| GCN | ~83.5 | ~83 |
| GAT | ~94.5 | ~94 |
The following content is from the initial cynricfu/MAGNN repo.
This repository provides a reference implementation of MAGNN as described in the paper:
MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding.
Xinyu Fu, Jiani Zhang, Ziqiao Meng, Irwin King.
The Web Conference, 2020.
Available at arXiv:2002.01680.
Recent versions of the following packages for Python 3 are required:
- PyTorch 1.2.0
- DGL 0.3.1
- NetworkX 2.3
- scikit-learn 0.21.3
- NumPy 1.17.2
- SciPy 1.3.1
Dependencies for the preprocessing code are not listed here.
The preprocessed datasets are available at:
The GloVe word vectors are obtained from GloVe. Here is the direct link for the version we used in DBLP preprocessing.
- Create
checkpoint/anddata/preprocesseddirectories - Extract the zip file downloaded from the section above to
data/preprocessed- E.g., extract the content of
IMDB_processed.ziptodata/preprocessed/IMDB_processed
- E.g., extract the content of
- Execute one of the following three commands from the project home directory:
python run_IMDB.pypython run_DBLP.pypython run_LastFM.py
For more information about the available options of the model, you may check by executing python run_IMDB.py --help
If you find MAGNN useful in your research, please cite the following paper:
@inproceedings{fu2020magnn,
title={MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding},
author={Xinyu Fu and Jiani Zhang and Ziqiao Meng and Irwin King},
booktitle = {WWW},
year={2020}
}