https://github.com/borgwardtlab/ggme
Official repository for the ICLR 2022 paper "Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions" https://openreview.net/forum?id=tBtoZYKd9n
https://github.com/borgwardtlab/ggme
evaluation-framework evaluation-metrics generative-model graph-learning machine-learning
Last synced: 3 months ago
JSON representation
Official repository for the ICLR 2022 paper "Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions" https://openreview.net/forum?id=tBtoZYKd9n
- Host: GitHub
- URL: https://github.com/borgwardtlab/ggme
- Owner: BorgwardtLab
- License: bsd-3-clause
- Created: 2022-02-25T08:52:18.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2022-07-05T18:45:03.000Z (over 3 years ago)
- Last Synced: 2023-06-09T12:05:13.346Z (over 2 years ago)
- Topics: evaluation-framework, evaluation-metrics, generative-model, graph-learning, machine-learning
- Language: Python
- Homepage: https://openreview.net/pdf?id=tBtoZYKd9n
- Size: 348 KB
- Stars: 12
- Watchers: 5
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ggme: Graph Generative Model Evaluation
This is the official repository for the ICLR 2022 paper "Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions" https://openreview.net/forum?id=tBtoZYKd9n
# Dependencies
Dependencies are managed using `poetry.` To setup the environment,
please run `poetry install` from the main directory (assuming the user
already has installed `poetry`).
# Running ggme
The primary script is contained in `main.py`. We assume that the user
has two distributions which they would like to compare using MMD, given
a specified kernel and descriptor function.
We assume that each distribution of graphs is stored as a list of `networkx`
graphs.
# Example script
We provide an example run in `main.py` based on predictions of a graph
generative model and the graphs in the corresponding test set. To run
this, execute the following code from the main directory.
```shell
cd src
poetry run python main.py
```
# Citing our work
Please consider citing our work:
```bibtex
@inproceedings{obray2022evaluation,
title = {Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions},
author = {Leslie O'Bray and Max Horn and Bastian Rieck and Karsten Borgwardt},
year = 2022,
booktitle = {International Conference on Learning Representations},
url = {https://openreview.net/forum?id=tBtoZYKd9n}
}
```