bisectgroup/ZINC_22

该数据集包含三个部分：训练集、验证集和测试集。每个部分都包含了名为smiles的字符串类型特征。训练集包含大约14.86亿个示例，大小为87870亿字节；验证集包含大约299.92万个示例，大小为1.77亿字节；测试集包含大约299.91万个示例，大小为1.78亿字节。整个数据集的下载大小为41.55亿字节，总大小为88.23亿字节。

The dataset consists of three parts: training set, validation set, and test set. Each part contains a feature named smiles of string type. The training set includes approximately 1.486 billion examples, with a size of 878.70 billion bytes; the validation set includes approximately 2.999 million examples, with a size of 177.48 million bytes; the test set includes approximately 2.999 million examples, with a size of 178.63 million bytes. The total download size of the dataset is 41.55 billion bytes, and the total size is 88.23 billion bytes.