five

Effects of perturbation order and basis set on alchemical predictions within 14 electron diatomic molecules series, detailed analysis of the various sources of errors

收藏
DataCite Commons2026-03-12 更新2026-05-04 收录
下载链接:
https://archive.materialscloud.org/doi/10.24435/materialscloud:qb-70
下载链接
链接失效反馈
官方服务:
资源简介:
This Dataset contains Supplementary information to the article "Effects of perturbation order and basis set on alchemical predictions" by Giorgio Domenichini, Guido Falk von Rudorff and O. Anatole von Lilienfeld. Our numerical analysis of potential energy estimates, and resulting binding curves, is based on CCSD reference results, and is limited to all neutral diatomics with 14 electrons (AlH ... N2). From those data can be predicted binding energy, equilibrium distance, and vibrational frequencies of neighboring out-of-sample diatomics with near CCSD quality using perturbations up to 5th order. The data stored as Comma Separated Value files (.csv) contain all the predictions within the 14 electron diatomics series, those are referred to the sections IIIA-IIIE of the article. To every file corresponds one of the eight basis sets used. The data are stored using consistently atomic units: Bohrs for length, electrons for charge and Hartrees for energy. For a better visualization using Jupyter notebooks and Pandas library the notebook "Display_data.ipynb" already contains the code to open the data as pandas' data frames.

本数据集为Giorgio Domenichini、Guido Falk von Rudorff与O. Anatole von Lilienfeld发表的论文《Effects of perturbation order and basis set on alchemical predictions》(扰动阶数与基组对炼金术预测的影响)的补充信息。 本文针对势能估计与所得结合曲线的数值分析,基于耦合簇单双激发(CCSD)参考结果,仅涵盖所有含14个电子的中性双原子分子(范围为AlH……N₂)。 借助最高至五阶的微扰方法,可基于上述数据预测接近CCSD精度的非训练集相邻双原子分子的结合能、平衡键长与振动频率。 以逗号分隔值(Comma Separated Value, .csv)文件存储的数据,包含14电子双原子分子系列的全部预测结果,对应论文的第IIIA至IIIE章节。 每个文件对应所使用的8种基组之一。 所有数据统一采用原子单位制:长度单位为玻尔(Bohr)、电荷单位为电子、能量单位为哈特里(Hartree)。 为便于使用Jupyter记事本(Jupyter notebooks)与Pandas库进行可视化,配套的"Display_data.ipynb"文件已包含将数据加载为Pandas数据框的代码。
提供机构:
Materials Cloud
创建时间:
2025-06-24
二维码
社区交流群
二维码
科研交流群
商业服务