CrySPY: a crystal structure prediction tool accelerated by machine learning
Tomoki Yamashita, Shinichi Kanehira, Nobuya Sato, Hiori Kino, Kei Terayama, Hikaru Sawahata, Takumi Sato, Futoshi Utsuno, Koji Tsuda, Takashi Miyake, Tamio Oguchi
Abstract
We have developed an open-source software called CrySPY, which is a crystal structure prediction tool written in Python 3, and runs on Unix/Linux platforms. CrySPY enables anyone to easily perform crystal structure prediction simulations for materials discovery and design, and automates structure generation, structure optimization, energy evaluation, and efficiently selecting candidates using machine learning. Several searching algorithms are available such as random search, evolutionary algorithm, Bayesian optimization, and Look Ahead based on Quadratic Approximation. Machine learning is employed to efficiently select candidates for priority optimization. CrySPY does not require complex machine learning techniques for users. In the latest version of CrySPY, both atomic and molecular random structures can be generated. CrySPY supports VASP, QUANTUM ESPRESSO, OpenMX, soiap, and LAMMPS for local structure optimization and energy evaluation. CrySPY is distributed under the MIT license at https://github.com/Tomoki-YAMASHITA/CrySPY. Documentation of CrySPY is also available at https://Tomoki-YAMASHITA.github.io/CrySPY_doc.