Abstract
We present PyXtal_FF—a package based on Python programming language—for developing machine learning potentials (MLPs). The aim of PyXtal_FF is to promote the application of atomistic simulations through providing several choices of atom-centered descriptors and machine learning regressions in one platform. Based on the given choice of descriptors (including the atom-centered symmetry functions, embedded atom density, SO4 bispectrum, and smooth SO3 power spectrum), PyXtal_FF can train MLPs with either generalized linear regression or neural network models, by simultaneously minimizing the errors of energy/forces/stress tensors in comparison with the data from ab-initio simulations. The trained MLP model from PyXtal_FF is interfaced with the Atomic Simulation Environment (ASE) package, which allows different types of light-weight simulations such as geometry optimization, molecular dynamics simulation, and physical properties prediction. Finally, we will illustrate the performance of PyXtal_FF by applying it to investigate several material systems, including the bulk SiO2, high entropy alloy NbMoTaW, and elemental Pt for general purposes. Full documentation of PyXtal_FF is available at https://pyxtal-ff.readthedocs.io.
Highlights
Molecular dynamics (MD) simulations have been used routinely to model the physical behaviors of many complex systems [1, 2, 3]
PyXtal FF creates machine learning potentials (MLPs) based on atom-centered descriptors such as atom-centered symmetry functions [11], embedded atom density [38], SO(4) bispectrum coefficients [19], and smooth SO(3) power spectrum [20]
The examples to be investigated mainly differ by the source of datasets, including (1) single SiO2 from pure MD simulation; (2) collective data set of NbMoTaW from various approaches; (3) elemental Pt consisting of bulk, surfaces and clusters from different runs of MD simulations
Summary
Molecular dynamics (MD) simulations have been used routinely to model the physical behaviors of many complex systems [1, 2, 3]. Several software packages [16, 29, 30, 31, 32, 33] were developed to train the MLPs. Among these, the RuNNer [16] is a closed source software for developing NNP, and ænet [34] is mainly written in FORTRAN/C and utilizes atom-centered symmetry functions (see section 2.1.1) as the descriptor. Our recent works suggested that NNP can be developed using bispectrum and power spectrum components as the descriptor while training on energy, forces, and stress simultaneously [35, 36]. PyXtal FF creates MLP based on atom-centered descriptors such as (weighted) atom-centered symmetry functions [11], embedded atom density [38], SO(4) bispectrum coefficients [19], and smooth SO(3) power spectrum [20]. We will demonstrate the usage of the current features of the package with SiO2 [30], high entropy alloy [39], and elemental Pt [37] as examples
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
