zincware / IPSuite

Machine Learned Interatomic Potential Tools
https://ipsuite.readthedocs.io
Eclipse Public License 2.0
19 stars 11 forks source link
machine-learning molecular-dynamics physics zntrack

PyTest ZnTrack zincware Documentation Status DOI PyPI version

IPS - The Inter Atomic Potential Suite

Logo

IPS provides you with tools to generate Machine Learned Interatomic Potentials. You can find the documentation at https://ipsuite.readthedocs.io

Install the package to get started or check out an interactive notebook Binder

pip install ipsuite

IPSuite relies on third-party ML packages. As these often come with different, sometimes incompatible requirements, an ipsuite installation is barebones. If you want to run your favorite ML code but are encountered with an import error, please install the package manually. You can look at the pyproject.toml to find the packages ipsuite is tested against.

Examples can be found at:

Docker Image

We provide an IPSuite docker image for Linux that includes the apax, mace and gap MLPs. You can use IPSuite directly from within the image by calling:

docker run -it -v "$(pwd):/app" --gpus all pythonf/ipsuite dvc repro
docker run -it -v "$(pwd):/app" --gpus all pythonf/ipsuite python
docker run -it -v "$(pwd):/app" --gpus all pythonf/ipsuite zntrack list
docker run -it -v "$(pwd):/app" --gpus all --rm -p 8888:8888 pythonf/ipsuite jupyter lab --ip=0.0.0.0 --port=8888 --allow-root

Fix Permission Issues

Running dvc repro via the docker container will create files owned by root:root. If you solely use docker this will not cause any issues. If you switch between docker and a dvc version on your host system, you might encounter permission errors. You can resolve them, by changing the ownership of the files. You can do this via the host chown "$(id -u):$(id -g)" -R . or from inside the docker container:

echo $(id -u):$(id -g)
docker run -it -v "$(pwd):/app" pythonf/ipsuite /bin/bash
addgroup --gid $GROUP_ID user
adduser --disabled-password --gecos '' --uid $USER_ID --gid $GROUP_ID user
chown user:user -R .

References

If you use IPSuite in your research and find it helpful please consider citing us.

@article{zillsCollaborationMachineLearnedPotentials2024,
  title = {Collaboration on {{Machine-Learned Potentials}} with {{IPSuite}}: {{A Modular Framework}} for {{Learning-on-the-Fly}}},
  shorttitle = {Collaboration on {{Machine-Learned Potentials}} with {{IPSuite}}},
  author = {Zills, Fabian and Schäfer, Moritz René and Segreto, Nico and Kästner, Johannes and Holm, Christian and Tovey, Samuel},
  date = {2024-04-03},
  journaltitle = {The Journal of Physical Chemistry B},
  shortjournal = {J. Phys. Chem. B},
  publisher = {American Chemical Society},
  issn = {1520-6106},
  doi = {10.1021/acs.jpcb.3c07187},
}

@misc{zillsZnTrackDataCode2024,
  title = {{{ZnTrack}} -- {{Data}} as {{Code}}},
  author = {Zills, Fabian and Sch{\"a}fer, Moritz and Tovey, Samuel and K{\"a}stner, Johannes and Holm, Christian},
  year = {2024},
  eprint={2401.10603},
  archivePrefix={arXiv},
}