PolyatomicComplexes

📖 Documentation

🚀 Installation

Using pip

1. Ensure you have python == 3.11.11 and set up a virtual environment.

pip install virtualenv
virtualenv .env --python=python3.11.11
source .env/bin/activate

2. Run the following

pip install -U polyatomic-complexes==1.0.7

Note: If you are having trouble with the environment setup please see the following demo in colab:

Using the repo

1. Clone the repo.

2. Ensure you have python >= 3.11.11 and set up a virtual environment.

pip install virtualenv
virtualenv .env --python=python3.11.11
source .env/bin/activate

3. Install the relevant packages.

For standard/minimal usage:

pip install -Ur requirements/requirements.txt

For graph based experiments:

pip install -Ur requirements/requirements_graph.txt

For materials based experiments:

pip install -Ur requirements/requirements_mat.txt

4. Get all large files from git lfs

git lfs fetch --all
git lfs pull

📜 License

This project is licensed under the MIT License.

🔬 Reference

@misc{khorana2024polyatomiccomplexestopologicallyinformedlearning,
      title={Polyatomic Complexes: A topologically-informed learning representation for atomistic systems}, 
      author={Rahul Khorana and Marcus Noack and Jin Qian},
      year={2024},
      eprint={2409.15600},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/2409.15600}, 
}