colabfit/Massive_Atomic_Diversity_MAD_train
收藏Hugging Face2025-10-16 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/colabfit/Massive_Atomic_Diversity_MAD_train
下载链接
链接失效反馈官方服务：
资源简介：
Massive Atomic Diversity MAD train是一个训练用的数据集，它通过扭曲稳定结构集来构建原子多样性。这个数据集虽然结构数量少于100k，但已经能够训练出与基于传统数据集训练的模型相竞争的通用原子间势。数据集包含了能量、原子力、Cauchy应力等属性，并包含了多种元素。
The Massive Atomic Diversity MAD train is a training dataset constructed by distorting stable structure sets to achieve atomic diversity. Despite having fewer than 100k structures, it has been shown to enable the training of universal interatomic potentials that are competitive with models trained on traditional datasets with two to three orders of magnitude more structures. The dataset includes properties such as energy, atomic forces, and Cauchy stress, and contains a variety of elements.
提供机构：
colabfit
