Machine learning models predict calculation outcomes with the transferability necessary for computational catalysis

data files, including ML models of dynamic classifiers, trajectories of electronic structure and geometric features, optimized geometries, and final csv files.