Structure featurization
Note
Structure featurization makes use of ænet’s compiled generate.x
trnset2ASCII.x tools. Make sure to install ænet and configure
the paths as described in Installation & Set-up.
Note
Alternative: For a pure Python/PyTorch implementation that does not require Fortran, see PyTorch-Based Featurization. The PyTorch implementation provides identical results with GPU acceleration support.
aenet-python can be used to featurize atomic environments with the
expansion method by Artrith et al. [1,2,3]. Local atomic environment
features can, furthermore, be combined to atomic structure features with
the approach by Gharakhanyan et al. [4].
[1] N. Artrith and A. Urban, Comput. Mater. Sci. 114, 2016, 135-150 (link4).
[2] N. Artrith, A. Urban, and G. Ceder, Phys. Rev. B 96, 2017, 014112 (link1).
[3] A. M. Miksch, T. Morawietz, J. Kästner, A. Urban, N. Artrith, Mach. Learn.: Sci. Technol. 2, 2021, 031001 (link2).
[4] V. Gharakhanyan, M. S. Aalto, A. Alsoulah, N. Artrith, A. Urban, ICLR 2023 (link3)
Example notebooks
Jupyter notebooks with examples how to use the featurization methods can be found in the notebooks directory within the repository.