colabfit/MP-ALOE
收藏资源简介:
--- configs: - config_name: default data_files: "co/*.parquet" - config_name: info data_files: "ds.parquet" license: cc-by-4.0 tags: - molecular dynamics - mlip - interatomic potential pretty_name: MP-ALOE --- ### <details><summary>Cite this dataset </summary>Kuner, M. C., Kaplan, A. D., Persson, K. A., Asta, M., and Chrzan, D. C. _MP-ALOE_. ColabFit, 2025. https://doi.org/None</details> #### This dataset has been curated and formatted for the ColabFit Exchange #### This dataset is also available on the ColabFit Exchange: https://materials.colabfit.org/id/DS_k14lrwu6r2mo_0 #### Visit the ColabFit Exchange to search additional datasets by author, description, element content and more. https://materials.colabfit.org <br><hr> # Dataset Name MP-ALOE ### Description MP-ALOE is a dataset of nearly 1 million DFT calculations computed with the r2SCAN meta-generalized gradient approximation, covering 89 elements. The dataset was constructed using active learning via Query by Committee (QBC) and downsampling via the DIRECT method, and primarily consists of off-equilibrium structures. Initial structures were generated by elemental substitution into prototype structures from the ICSD and Materials Project databases (restricted to 2-8 atoms and up to ternary compositions). QBC used an ensemble of interatomic potentials (initially MACE-MP-0, CHGNet, and M3GNet, followed by iteratively trained MACE models) to select structures with energy uncertainty exceeding 100 meV/atom, force uncertainty exceeding 100 meV/Å, or stress uncertainty exceeding 100 meV/ų. DIRECT downsampling reduced approximately 500,000 selected structures to approximately 125,000 for DFT calculation. Near-equilibrium structures from the Materials Project (up to 3 elements, up to 32 atoms, approximately 30,000 structures) were recalculated with identical DFT settings. A two-stage VASP workflow was applied: an initial static calculation using PBE, followed by r2SCAN relaxation for three ionic steps. In total, 909,792 frames from 303,264 structure relaxations are included. DFT calculations used projector-augmented wave (PAW) potentials, a 680 eV plane-wave cutoff, and KSPACING=0.2, with additional parameters from the MP24RelaxSet in pymatgen. Calculations were managed by the atomate2 workflow package. ### Dataset authors Matthew C. Kuner, Aaron D. Kaplan, Kristin A. Persson, Mark Asta, Daryl C. Chrzan ### Publication https://doi.org/10.1038/s41524-025-01834-9 ### Original data link https://figshare.com/articles/dataset/MP_ALOE_An_r2SCAN_dataset_for_universal_machine_learning_interatomic_potentials/29452190 ### License CC-BY-4.0 ### Number of unique molecular configurations 909789 ### Number of atoms 5891262 ### Elements included Ac, Ag, Al, Ar, As, Au, B, Ba, Be, Bi, Br, C, Ca, Cd, Ce, Cl, Co, Cr, Cs, Cu, Dy, Er, Eu, F, Fe, Ga, Gd, Ge, H, He, Hf, Hg, Ho, I, In, Ir, K, Kr, La, Li, Lu, Mg, Mn, Mo, N, Na, Nb, Nd, Ne, Ni, Np, O, Os, P, Pa, Pb, Pd, Pm, Pr, Pt, Pu, Rb, Re, Rh, Ru, S, Sb, Sc, Se, Si, Sm, Sn, Sr, Ta, Tb, Tc, Te, Th, Ti, Tl, Tm, U, V, W, Xe, Y, Yb, Zn, Zr ### Properties included energy, atomic forces, electronic band gap, cauchy stress <br> <hr> # Usage - `ds.parquet` : Aggregated dataset information. - `co/` directory: Configuration rows each include a structure, calculated properties, and metadata. - `cs/` directory : Configuration sets are subsets of configurations grouped by some common characteristic. If `cs/` does not exist, no configurations sets have been defined for this dataset. - `cs_co_map/` directory : The mapping of configurations to configuration sets (if defined). <br> #### ColabFit Exchange documentation includes descriptions of content and example code for parsing parquet files: - [Parquet parsing: example code](https://materials.colabfit.org/docs/how_to_use_parquet) - [Dataset info schema](https://materials.colabfit.org/docs/dataset_schema) - [Configuration schema](https://materials.colabfit.org/docs/configuration_schema) - [Configuration set schema](https://materials.colabfit.org/docs/configuration_set_schema) - [Configuration set to configuration mapping schema](https://materials.colabfit.org/docs/cs_co_mapping_schema)
配置项: - 配置名称:default 数据文件:"co/*.parquet" - 配置名称:info 数据文件:"ds.parquet" 许可证:cc-by-4.0 标签: - 分子动力学 - 机器学习原子间势(MLIP) - 原子间势 友好显示名称:MP-ALOE --- ### <details><summary>引用本数据集</summary>Kuner, M. C., Kaplan, A. D., Persson, K. A., Asta, M., 及 Chrzan, D. C. _MP-ALOE_. ColabFit, 2025. https://doi.org/None</details> #### 本数据集已完成整理与格式化,适配ColabFit交换平台 #### 本数据集亦可在ColabFit交换平台获取: https://materials.colabfit.org/id/DS_k14lrwu6r2mo_0 #### 访问ColabFit交换平台,可按作者、数据集描述、元素组成等维度检索更多数据集: https://materials.colabfit.org <br><hr> # 数据集名称 MP-ALOE ### 数据集描述 MP-ALOE 是包含近百万份密度泛函理论(Density Functional Theory, DFT)计算结果的数据集,采用 r2SCAN 元广义梯度近似(meta-generalized gradient approximation, meta-GGA)方法完成计算,覆盖89种化学元素。本数据集通过基于委员会查询(Query by Committee, QBC)的主动学习方法构建,并通过 DIRECT 方法进行下采样,主体包含非平衡结构。初始结构通过将元素替换自无机晶体结构数据库(Inorganic Crystal Structure Database, ICSD)与材料项目(Materials Project)数据库中的原型结构生成,限制条件为2~8个原子、最多三元组分。QBC 采用原子间势集成模型(初始为 MACE-MP-0、CHGNet 及 M3GNet,后续为迭代训练的 MACE 模型),选取能量不确定度超过100 meV/原子、力不确定度超过100 meV/Å 或应力不确定度超过100 meV/ų 的结构。DIRECT 下采样将约50万个选中的结构缩减至约12.5万个用于 DFT 计算。从材料项目数据库获取的近平衡结构(最多3种元素、最多32个原子,共约3万个结构)采用相同的 DFT 设置重新计算。本数据集采用两阶段 VASP 工作流:首先使用 PBE 泛函进行静态计算,随后以 r2SCAN 泛函完成三步离子弛豫。本数据集共包含来自303264次结构弛豫的909792个构型。DFT 计算采用投影缀加波(Projector Augmented Wave, PAW)势、680 eV 平面波截断能、KSPACING=0.2,并使用 pymatgen 中的 MP24RelaxSet 其余参数。计算工作流由 atomate2 软件包管理。 ### 数据集作者 马修·C·库纳(Matthew C. Kuner)、亚伦·D·卡普兰(Aaron D. Kaplan)、克里斯汀·A·佩尔松(Kristin A. Persson)、马克·阿斯塔(Mark Asta)、达里尔·C·赫赞(Daryl C. Chrzan) ### 发表文献 https://doi.org/10.1038/s41524-025-01834-9 ### 原始数据链接 https://figshare.com/articles/dataset/MP_ALOE_An_r2SCAN_dataset_for_universal_machine_learning_interatomic_potentials/29452190 ### 许可证 CC-BY-4.0 ### 唯一分子构型数量 909789 ### 原子总数 5891262 ### 包含元素 锕(Ac)、银(Ag)、铝(Al)、氩(Ar)、砷(As)、金(Au)、硼(B)、钡(Ba)、铍(Be)、铋(Bi)、溴(Br)、碳(C)、钙(Ca)、镉(Cd)、铈(Ce)、氯(Cl)、钴(Co)、铬(Cr)、铯(Cs)、铜(Cu)、镝(Dy)、铒(Er)、铕(Eu)、氟(F)、铁(Fe)、镓(Ga)、钆(Gd)、锗(Ge)、氢(H)、氦(He)、铪(Hf)、汞(Hg)、钬(Ho)、碘(I)、铟(In)、铱(Ir)、钾(K)、氪(Kr)、镧(La)、锂(Li)、镥(Lu)、镁(Mg)、锰(Mn)、钼(Mo)、氮(N)、钠(Na)、铌(Nb)、钕(Nd)、氖(Ne)、镍(Ni)、镎(Np)、氧(O)、锇(Os)、磷(P)、镤(Pa)、铅(Pb)、钯(Pd)、钷(Pm)、镨(Pr)、铂(Pt)、钚(Pu)、铷(Rb)、铼(Re)、铑(Rh)、钌(Ru)、硫(S)、锑(Sb)、钪(Sc)、硒(Se)、硅(Si)、钐(Sm)、锡(Sn)、锶(Sr)、钽(Ta)、铽(Tb)、锝(Tc)、碲(Te)、钍(Th)、钛(Ti)、铊(Tl)、铥(Tm)、铀(U)、钒(V)、钨(W)、氙(Xe)、钇(Y)、镱(Yb)、锌(Zn)、锆(Zr) ### 包含属性 能量、原子受力、电子带隙、柯西应力 <br> <hr> # 使用说明 - `ds.parquet`:聚合数据集信息文件。 - `co/` 目录:每个配置行均包含结构、计算得到的属性及元数据。 - `cs/` 目录:构型集为按共同特征分组的构型子集。若`cs/`目录不存在,则未为本数据集定义任何构型集。 - `cs_co_map/` 目录:构型与构型集的映射关系(若已定义)。 <br> #### ColabFit 交换平台文档包含数据集内容说明与 Parquet 文件解析示例代码: - [Parquet 文件解析示例代码](https://materials.colabfit.org/docs/how_to_use_parquet) - [数据集信息 Schema](https://materials.colabfit.org/docs/dataset_schema) - [构型 Schema](https://materials.colabfit.org/docs/configuration_schema) - [构型集 Schema](https://materials.colabfit.org/docs/configuration_set_schema) - [构型集与构型映射关系 Schema](https://materials.colabfit.org/docs/cs_co_mapping_schema)



