# DScribe: Library of Descriptors for Machine Learning in Materials Science

## TL;DR

DScribe is an open-source software library that provides a collection of feature descriptors for atomistic materials simulations, facilitating machine learning applications in materials science.

## Contribution

It offers a user-friendly, ready-to-use set of descriptors for atomistic data, streamlining the integration of machine learning in materials research.

## Key findings

- Accelerates machine learning workflows for materials property prediction
- Demonstrates application in formation energy and ionic charge prediction
- Provides multiple descriptors including Coulomb matrix and SOAP

## Abstract

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.