GGFM is an open-source pipeline for graph foundation model based on PyTorch. We integrate SOTA graph foundation models.
The relevant documentation is available here. It is under development, welcome join us!
GGFM works with the following operating systems:
- Linux
1. Python environment (Optional): We recommend using Conda package manager
conda create -n ggfm python=3.8
source activate ggfm2. Pytorch: Follow their tutorial to run the proper command according to your OS and CUDA version. For example:
pip install torch torchvision torchaudio3. DGL: Follow their tutorial to run the proper command according to your OS and CUDA version. For example:
pip install dgl -f https://data.dgl.ai/wheels/repo.html4. PyG: Follow their tutorial to run the proper command according to your OS and CUDA version. For example:
pip install torch_geometric4. Install GGFM:
- install from pypi
pip install ggfm- install from source
git clone https://github.com/BUPT-GAMMA/ggfm
cd ggfm
pip install .