This repository contains code and data for the dissertation "Machine Learning for Atomistic Modeling: Representations and Thermal Transport" by Marcel F. Langer.
This repository is available at https://github.com/sirmarcel/doctor-archive. Selected versions are archived on Zenodo, under doi:10.5281/zenodo.7963151. The official version of the thesis can be found at doi:10.14279/depositonce-18647, the website for the thesis at marcel.science/doctor.
Each subfolder in this repository contains a README.md with additional information. The subfolders are:
results/: Additional data that was not previously publishedplots/: Plottings scriptsthesis/: LaTeX source for the thesiswork/: Training data and workflows that were not previously published (silicon experiment)meta/: Scripts for assembling this archive; can be ignored but is retained for transparency.
The plots are largely based on data published previously in:
- https://github.com/sirmarcel/gknet-archive (doi:10.5281/zenodo.7767432)
- https://github.com/sirmarcel/glp-archive (doi:10.5281/zenodo.7852530)
Any additional data required by this thesis can be found in this repository.
v1.1: added silicon datav1.0: first publication doi:10.5281/zenodo.7963152
This repository is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license.