Surrogate Theory Accelerated Line-search Kit (STALK) is a Python implementation of The Surrogate Hessian Accelerated Parallel Line-search method. The method is intended for optimizing and performing energy minimization of atomic structures in the presence of statistical noise.
Note! The repository was renamed 'surrogate_hessian_relax'->'stalk' on 19 Dec 2025, and the code usage has changed substantially upon python packaging. To complete projects in the old code base, keep using v0.1 or reach out for help in migration. We regret any inconvenience and anticipate fewer breaking changes in the future.
Upon publishing results based on the method, we kindly ask you to cite The original work.
Juha Tiihonen, Paul R. C. Kent, and Jaron T. Krogel
The Journal of Chemical Physics
156, 054104 (2022)
Installation instructions, examples and other information can be found in the online Documentation.
The software and its documentation are under development with no warranties. Support may be inquired by contacting the authors.
The authors of this method are Juha Tiihonen, Paul R. C. Kent and Jaron T. Krogel, working in the Center for Predictive Simulation of Functional Materials (https://cpsfm.ornl.gov/)
This work has been authored in part by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE).