DScribe: Library of descriptors for machine learning in materials science

DScribe is a software package for machine learning that provides popular feature transformations (“descriptors”) for atomistic materials simulations. DScribe accelerates the application of machine learning for atomistic property prediction by providing user-friendly, off-the-shelf descriptor implementations. The package currently contains implementations for Coulomb matrix, Ewald sum matrix, sine matrix, Many-body Tensor Representation (MBTR), Atom-centered Symmetry Function (ACSF) and Smooth Overlap of Atomic Positions (SOAP). Usage of the package is illustrated for two different applications: formation energy prediction for solids and ionic charge prediction for atoms in organic molecules. The package is freely available under the open-source Apache License 2.0.DScribe是一款面向机器学习的软件包，专为原子尺度材料模拟提供主流的特征变换（即描述符，descriptors）。该软件包通过提供易用的开箱即用型描述符实现方案，加速了机器学习在原子性质预测领域的应用。目前该软件包已实现以下描述符：库仑矩阵（Coulomb matrix）、埃瓦尔德求和矩阵（Ewald sum matrix）、正弦矩阵（sine matrix）、多体张量表示（Many-body Tensor Representation，MBTR）、原子中心对称函数（Atom-centered Symmetry Function，ACSF）以及原子位置平滑重叠（Smooth Overlap of Atomic Positions，SOAP）。本文以两类应用场景演示了该软件包的使用方法：固体的形成能预测，以及有机分子中原子的离子电荷预测。该软件包遵循开源Apache许可证2.0（Apache License 2.0）条款，可免费获取使用。