nimashoghi/salex

该数据集包含多个特征字段，如位置（pos）、单元（cell）、原子序数（atomic_numbers）、原子数量（natoms）、标签（tags）、样本ID（sid）、能量（energy）、力（forces）、应力（stress）、固定原子（fixed）、周期性边界条件（pbc）和文件ID（fid）。数据集分为训练集和验证集，训练集包含10,447,765个样本，验证集包含553,218个样本。数据集总大小为7,028,386,987字节，下载大小为3,962,818,586字节。

The dataset includes multiple feature fields such as position (pos), cell (cell), atomic numbers (atomic_numbers), number of atoms (natoms), tags (tags), sample ID (sid), energy (energy), forces (forces), stress (stress), fixed atoms (fixed), periodic boundary conditions (pbc), and file ID (fid). The dataset is divided into a training set and a validation set, with the training set containing 10,447,765 samples and the validation set containing 553,218 samples. The total size of the dataset is 7,028,386,987 bytes, and the download size is 3,962,818,586 bytes.