iknow-lab/open-materials-guide-0210-embeddings

该数据集包含四个字段：id、contribution、recipe和contributions_embedding。id是字符串类型，用于唯一标识每个记录；contribution是字符串类型，可能包含某种贡献信息；recipe也是字符串类型，可能包含某种配方信息；contributions_embedding是一个浮点数序列，可能表示对贡献信息的向量嵌入。数据集分为三个部分：训练集、标准影响测试集和高度影响测试集，分别包含16026、1472和169个示例。数据集的总大小为482,572,854字节。

The dataset includes four fields: id, contribution, recipe, and contributions_embedding. The id is a string type used to uniquely identify each record; contribution is a string type that may contain some contribution information; recipe is also a string type that may contain some recipe information; contributions_embedding is a sequence of floating-point numbers that may represent vector embeddings of contribution information. The dataset is divided into three parts: training set, standard impact test set, and high impact test set, containing 16,026, 1,472, and 169 examples respectively. The total size of the dataset is 482,572,854 bytes.