ICET – A Python Library for Constructing and Sampling Alloy Cluster Expansions

Abstract

Alloy cluster expansions (CEs) provide an accurate and computationally efficient mapping of the potential energy surface of multi‐component systems that enables comprehensive sampling of the many‐dimensional configuration space. Here, integrated cluster expansion toolkit (ICET), a flexible, extensible, and computationally efficient software package, is introduced for the construction and sampling of CEs. ICET is largely written in Python for easy integration in comprehensive workflows, including first‐principles calculations for the generation of reference data and machine learning libraries for training and validation. The package enables training using a variety of linear regression algorithms with and without regularization, Bayesian regression, feature selection, and cross‐validation. It also provides complementary functionality for structure enumeration and mapping as well as data management and analysis. Potential applications are illustrated by two examples, including the computation of the phase diagram of a prototypical metallic alloy and the analysis of chemical ordering in an inorganic semiconductor.

Efficient construction of linear models in materials modeling and applications to force constant expansions

“…While the methods achieve comparable RMSE scores, the optimal LASSO solution contains a much larger number of features. The tendency of (standard) LASSO to over-select is known 35 and we have observed this behavior also in other applications such as alloy cluster expansions 47 . A physical understanding of the shortcoming of LASSO is obtained by inspecting the FCs directly (Fig.…”

A Library of Late Transition Metal Alloy Dielectric Functions for Nanophotonic Applications

“…The DFs were calculated with enumerated structures with up to four atoms per primitive cell, as well as SQSs with 20–24 atoms (depending on target concentration). [ 32,54,55 ] The latter were generated with a simulated annealing approach minimizing the deviation from the random limit of the cluster correlation of pairs shorter than 3.25 a and triplets shorter than 1.5 a ( a being the lattice parameter), and with additional weight assigned to the cluster correlation of short‐ranged pairs, using the default parameters suggested in ref. [ 56 ] .…”

Atomistic understanding of the LiNiO2–NiO2 phase diagram from experimentally guided lattice models

“…We calculated the compositional phase diagram of Li x NiO 2 performing Monte Carlo simulations in the semi-grand canonical ensemble-i.e., at constant temperature T, total number of sites N and chemical potential difference Dm between vacancies and Li ions-with the aid of the mchammer module of ICET. 44 Simulations were started at 2000 K and equally spaced chemical potential differences between À0.6 eV per f.u. to 0.6 eV per f.u.…”