UnconvBench
收藏资源简介:
UnconvBench是由香港城市大学化学系等机构创建的一个综合数据集,旨在评估模型在非传统晶体材料上的预测性能。该数据集包含11个子数据集,涵盖了2D晶体、金属有机框架(MOF)、缺陷晶体等多种非传统晶体材料。数据集的创建过程涉及从Materials Project和JARVIS等数据库收集原始数据,并通过精心设计的数据处理步骤生成。UnconvBench主要应用于材料科学领域,特别是用于预测和发现新型非传统晶体材料的物理和化学性质。
UnconvBench is a comprehensive dataset developed by the Department of Chemistry at City University of Hong Kong and other institutions, aiming to evaluate the predictive performance of models on unconventional crystalline materials. This dataset includes 11 sub-datasets, covering a variety of unconventional crystalline materials such as 2D crystals, metal-organic frameworks (MOFs), defective crystals and others. The creation of UnconvBench involves collecting raw data from databases including Materials Project and JARVIS, and generating the final dataset through meticulously designed data processing procedures. UnconvBench is mainly applied in the field of materials science, particularly for predicting and discovering the physical and chemical properties of novel unconventional crystalline materials.




