Data for: Machine-learning-accelerated simulations enable heuristic-free surface reconstruction

This is the dataset for the publication "Machine-learning-accelerated simulations to enable automatic surface reconstruction", by X. Du, J.K. Damewood, J.R. Lunger, R. Millan, B. Yildiz, L. Li, and R. Gómez-Bombarelli. The repository contains the density-functional theory (DFT) data used to train the neural network force fields (NFF), selected results from our GaN(0001), Si(111), and SrTiO3(001) Virtual Surface Site Relaxation-Monte Carlo (VSSR-MC) runs, and Jupyter notebooks used for analysis and plots. To run the .ipynb's, you will need to install surface-sampling (tested up to commit 02820d339eed6291b6af6ccb809f154ad6244110 on master) and NeuralForceField (tested up to commit 72d1f32f43f202c1a466116beeed15845a6456e7 on master) from the Rafael Gómez-Bombarelli Group @ MIT.