Graph embedding is the construction of vector representations of graphs to perform tasks such as classification or clustering. This repo contains experiments in constructing new graph embeddings, including a sparse selected motif embedding.
pipenv can be used for dependency management.
To install the dependencies in a local virtual environment, run:
python -m pip install pipenv
export PIPENV_VENV_IN_PROJECT=true
pipenv installTo plot graphs the cairo 2D graphics library is required.
It can be installed on Linux by running
sudo apt install libcairo2-devWindows installation instructions can be found here.
To view and run the jupyter notebooks, run:
pipenv run jupyter notebook