ASSYST or Automated Small SYmmetric Structure Training

A flexible reference implementation of the ASSYST method to generate transferable training data for machine learning potentials.

ASSYST is the Automated Small Symmetric Structure Training, a training protocol, aimed at providing comprehensive, transferable training sets for machine learning interatomic potentials (MLIP) automatically. A detailed explanation and verification of the method can be found in our papers. 12 ASSYST gives up the notion of fitting potentials to individual phases or structures and instead tries to deliver a training set spanning the full potential energy surface (PES) of a material.

This software package is a minimal implementation of this idea, designed to be as flexible as possible without assuming either a specific MLIP, reference data, or workflow manager in mind. It is built on ASE and can use any of its calculators. It also assumes that you label its output structures with reference energies and forces on your own, either with an ASE calculator or by any other method. For a ready-to-run implementation that targets Atomic Cluster Expansion and Moment Tensor Potentials fit to Density Functional Theory (DFT) data check out pyiron_potentialfit.

Citation

Please use the following citation when referencing the method in your work.

@article{poul2025automated,
  title={Automated generation of structure datasets for machine learning potentials and alloys},
  volume={11},
  DOI={10.1038/s41524-025-01669-4},
  number={1},
  journal={npj Computational Materials},
  author={Poul, Marvin and Huber, Liam and Neugebauer, J\"org},
  year={2025},
  month={Jun}
}