Force field parameterization of azobenzene derivatives based on machine learning potential energy surface prediction and high-throughput fitting

All data and code supporting the findings of this study, including the QM datasets for ten AZDs, the GPR-interpolated PESs, the GBT/RF models for PES extrapolation, and the tabulated potentials for force field implementation

本研究各项发现所依托的全部数据与代码，涵盖十种AZD的量子力学（Quantum Mechanics）数据集、经高斯过程回归（Gaussian Process Regression, GPR）插值得到的势能面（Potential Energy Surface, PES）、用于势能面外推的梯度提升树（Gradient Boosting Tree, GBT）与随机森林（Random Forest, RF）模型，以及用于力场（Force Field）实现的制表势能。