PubChem Molecular Properties Dataset

该数据集来源于PubChem数据库，专注于小分子五大选定分子特性的构建。数据集包含了从小分子SMILES表示中提取的分子特性，如氢键受体计数、氢键供体计数、可旋转键计数、辛醇-水分配系数以及分子量。该数据集包含10,000个分子，并进一步划分为多个包含500,000个分子的文件。其任务是依据结构数据预测分子性质。

This dataset is sourced from the PubChem database, focusing on the construction of five selected molecular properties for small molecules. It includes molecular properties extracted from the SMILES representations of small molecules, such as the number of hydrogen bond acceptors, the number of hydrogen bond donors, the number of rotatable bonds, octanol-water partition coefficient, and molecular weight. The dataset contains 10,000 molecules in total, and is further divided into multiple files, each containing 500,000 molecules. The corresponding task is to predict molecular properties based on their structural data.