GraphGenerator

Toolkit for simulating observed graphs, generating new graphs and evaluating graph generators.

Installation

Environments

If users want to use some deep learning based graph generators, deep learning dependencies are required such as Pytorch or Tensorflow. We prefer to use PyTorch as dependency.

1. Install Pytorch

pip install torch==1.8.1+cu111 torchvision==0.9.1+cu111 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html

2. Clone and install

git clone https://github.com/xiangsheng1325/GraphGenerator.git
cd GraphGenerator
pip install -r requirements.txt
pip install -e .

Dependencies

Users need to install specific dependencies to support some graph generators, which is listed here:

Project organization

This project is modularized to benefit further contributions on it. Please organize this project according to following structure:

GraphGenerator/
|___GraphGenerator/  # source code
|   |___models/ # graph generator implementations
|   |   |___bigg/
|   |   |   |___tree_model.py
|   |   |   |___...
|   |   |___sbm.py
|   |   |___...
|   |___metrics/
|   |   |___mmd.py
|   |   |___...
|   |___train.py
|   |___...
|
|___setup.py 
|
|___config/  # detailed configurations of complex models
|   |___graphite.yaml
|   |___...
|
|___data/  # raw data / cooked data
|   |___google.txt
|   |___...
|
|___exp # trained model and generated graphs
|   |___VGAE/
|   |___...
|
|___...

GraphGenerator Usage

Here are some examples of using this toolkit.

1. Preprocess data

We prefer to converting graph data into the same data type. If the input data is ready, this step can be skipped.

Example:

• run python -m GraphGenerator --phase preprocessing -i google.txt -o google.graph

2. Test the usage of graph generator

Before training the deep learning based graph generators, we prefer to test whether there are bugs in our model implementations. If the generator runs well, this step can be skipped.

Example:

• run python -m GraphGenerator --phase test -g bigg --config config/bigg.yaml

Note that some algorithms may be affected by the CUDA version. (For example, Bigg may encounter problems during testing, please refer to this page to find resolutions.)

3. Train and infer new graphs

Enjoy your graph simulation and graph data generation.

Example:

• run python -m GraphGenerator --phase train -i google.graph -g vgae --config config/vgae.yaml

4. Evaluate the results

Calculating the distance between two set of graphs to evaluate the experimental results.

Example:

• run python -m GraphGenerator --phase evaluate -i new_google.graphs -r google.graph

Reference

Please use the following BibTex to cite this work if it makes contributions to your publications.

BibTex:

@Article{Xiang2021General,
    author={Xiang, Sheng and Wen, Dong and Cheng, Dawei and Zhang, Ying and Qin, Lu and Qian, Zhengping and Lin, Xuemin},
    title={General Graph Generators: Experiments, Analyses, and Improvements},
    url={https://doi.org/10.1007/s00778-021-00701-5},
    doi={10.1007/s00778-021-00701-5},
    journal={The VLDB Journal},
    publisher={Springer}
    issn={0949-877X},
    pages={1--29},
    year={2021},
    month={Oct},
    day={07},
}