trmteb/msmarco-tr

该数据集包含三种配置：corpus、default和queries。corpus配置包含文本数据和相关标识，default配置包含查询标识、文本标识和分数，queries配置包含查询文本和相关标识。数据集分为多个数据分片，corpus分片包含718217条文本数据，default配置分为训练集、验证集和测试集，分别包含426208、53276和53277条数据，queries分片包含501428条查询数据。

The dataset consists of three configurations: corpus, default, and queries. The corpus configuration includes text data and related identifiers, the default configuration includes query identifiers, text identifiers, and scores, and the queries configuration includes query texts and related identifiers. The dataset is divided into multiple data splits, with the corpus split containing 718217 text entries, the default configuration divided into training, validation, and test sets with 426208, 53276, and 53277 entries respectively, and the queries split containing 501428 query entries.