colabfit/discrepencies_and_error_metrics_NPJ_2023_enhanced_validation_set

该数据集是‘discrepencies and error metrics NPJ 2023 enhanced validation set’，包含了经过增强的验证集，由稀有事件的结构组成。它基于原始的mlearn_Si_train数据集进行了修改，用包含迁移间隙原子的结构替换了约54%的数据，以开发具有更好扩散性评分的模型。增强的验证集共有50个结构，包括从原始训练数据集替换的120个结构中随机选取的20个结构，来自AIMD模拟的11个空位稀有事件快照和19个间隙稀有事件快照。此外，还构建了包含单个迁移空位或间隙的原子构型的间隙稀有事件和空位稀有事件测试集，每个测试集包含100个来自1230 K下AIMD模拟的快照。数据集中还包括了附加的详细信息，存储在以dataset_为前缀的列中。

The dataset discrepencies and error metrics NPJ 2023 enhanced validation set contains an enhanced validation set composed of structures with rare events. It is based on the modification of the original mlearn_Si_train dataset, with approximately 54% of the data replaced by structures containing migrating interstitials to develop models with better diffusivity scores. The enhanced validation set consists of 50 structures, including 20 structures randomly selected from the 120 replaced structures of the original training dataset, 11 snapshots with vacancy rare events (RE) from AIMD simulations, and 19 snapshots with interstitial RE from AIMD simulations. Additionally, interstitial-RE and vacancy-RE testing sets are constructed, each consisting of 100 snapshots of atomic configurations with a single migrating vacancy or interstitial, respectively, from AIMD simulations at 1230 K. The dataset also includes additional details stored in columns prepended with dataset_.