hpprc/jsec

该数据集包含两种配置：default和sim。default配置包含英语和日语的字符串类型数据，适用于一般的语言处理任务。sim配置除了包含这两种语言的数据外，还包含一个表示余弦相似度的浮点数类型数据，适用于需要语言相似度分析的任务。两种配置都只有一个训练集分割，且提供了数据集的大小和下载大小的信息。

The dataset contains two configurations: default and sim. The default configuration includes string-type data in English and Japanese, suitable for general language processing tasks. The sim configuration, in addition to the data in these two languages, includes a float-type data representing cosine similarity, suitable for tasks requiring language similarity analysis. Both configurations have only one training set split and provide information on the size of the dataset and the download size.