DScribe: Library of descriptors for machine learning in materials science
DScribe: библиотека дескрипторов для машинного обучения в материаловедении
2019-09-26
SCID: 54.1/t6gh5jju
Discuss with AI
Coulomb matrixMany-body Tensor Representation (MBTR)Smooth Overlap of Atomic Positions (SOAP)atomistic descriptorsmaterials property prediction
Figures from the paper
Abstract (AI)
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. Program Title: DScribe Program Files doi: http://dx.doi.org/10.17632/vzrs8n8pk6.1 Licensing provisions: Apache-2.0 Programming language: Python/C/C++ Supplementary material: Supplementary Information as PDF Nature of problem: The application of machine learning for materials science is hindered by the lack of consistent software implementations for feature transformations. These feature transformations, also called descriptors, are a key step in building machine learning models for property prediction in materials science. Solution method: We have developed a library for creating common descriptors used in machine learning applied to materials science. We provide an implementation the following descriptors: Coulomb matrix, Ewald sum matrix, sine matrix, Many-body Tensor Representation (MBTR), Atom-centered Symmetry Functions (ACSF) and Smooth Overlap of Atomic Positions (SOAP). The library has a python interface with computationally intensive routines written in C or C++. The source code, tutorials and documentation are provided online. A continuous integration mechanism is set up to automatically run a series of regression tests and check code coverage when the codebase is updated.
Key Findings
1
DScribe addresses inconsistent descriptor software implementations by offering user-friendly, off-the-shelf feature transformations for atomistic property prediction.
2
DScribe includes online source code, tutorials, documentation, and continuous integration with regression tests and code-coverage checks.
3
DScribe provides an open-source Python library with C/C++ implementations of widely used atomistic descriptors for materials machine learning.
4
The library is demonstrated for formation-energy prediction in solids and ionic-charge prediction for atoms in organic molecules.
5
The package implements Coulomb matrix, Ewald sum matrix, sine matrix, MBTR, ACSF, and SOAP representations.
Research Object
DScribe software library for atomistic materials simulations
Research Subject
Implementation and provision of machine-learning descriptors for atomistic property prediction
Publication Details
Publication Date
2019-09-26
Journal
Publisher
ISSN
Cited by
853
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai10
Scikit-learn: Machine Learning in Python2012
Representation Learning: A Review and New Perspectives2013
Commentary: The Materials Project: A materials genome approach to accelerating materials innovation2013
Neural Message Passing for Quantum Chemistry2017
The atomic simulation environment—a Python library for working with atoms2017
Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules2018
On representing chemical environments2013
Quantum chemistry structures and properties of 134 kilo molecules2014
Comparing molecules and solids across structural and alchemical space2016
Machine learning for heterogeneous catalyst design and discovery2018
Cited by8
Recent advances and applications of machine learning in solid-state materials science2019
Machine Learning Interatomic Potentials as Emerging Tools for Materials Science2019
Performance and Cost Assessment of Machine Learning Interatomic Potentials2020
Self-Driving Laboratories for Chemistry and Materials Science2024
AI for Complex Catalytic Systems: High-Entropy Alloys in Electrocatalytic Acetylene Semihydrogenation2025
Machine Learning Electron Density Prediction Using Weighted Smooth Overlap of Atomic Positions2023
Advancing chemical engineering technology with artificial intelligence2025
Accurate Prediction of Three-Body Intermolecular Interactions via Electron Deformation Density-Based Machine Learning2023