mteb/NeuCLIR2023Retrieval_zho_top_250_only_w_correct

该数据集包含三个配置：文本数据集（corpus）、默认配置（default）和查询数据集（queries）。文本数据集包含_id和text两个字段，默认配置包含query-id、corpus-id和score三个字段，查询数据集包含_id和text两个字段。每个配置都有训练集划分，其中文本数据集的训练集大小为35271864.10211094字节，包含17161个示例。默认配置的训练集大小为376035字节，包含6837个示例。查询数据集的训练集大小为5592字节，包含76个示例。

The dataset consists of three configurations: a text corpus (corpus), a default configuration (default), and a query dataset (queries). The text corpus includes two fields: _id and text, while the default configuration includes query-id, corpus-id, and score. The query dataset also includes _id and text fields. Each configuration has a training set, with the text corpus training set being 35271864.10211094 bytes in size and containing 17161 examples. The default configuration training set is 376035 bytes and contains 6837 examples. The query dataset training set is 5592 bytes and contains 76 examples.