Litcius/Paper detail

Big data benchmarking: how do DFT methods across the rungs of Jacob's ladder perform for a dataset of 122k CCSD(T) total atomization energies?

Amir Karton

2024Physical Chemistry Chemical Physics13 citationsDOIOpen Access PDF

Abstract

Assesses the performance of DFT for atomization energies using a big-data set of 122 000 small drug-like molecules relative to CCSD(T) reference values. B3LYP emerges as the best performer (MAD = 4.1 kcal mol −1 ) followed by M06-L (MAD = 6.2 kcal mol −1 ).

Topics & Concepts

BenchmarkingComputer scienceComputational chemistryStatisticsMathematicsChemistryManagementEconomicsMachine Learning in Materials ScienceAdvanced Chemical Physics StudiesPhase Equilibria and Thermodynamics
Big data benchmarking: how do DFT methods across the rungs of Jacob's ladder perform for a dataset of 122k CCSD(T) total atomization energies? | Litcius