mixedbread-ai/vidore-shiftproject_test

该数据集包含三个配置：corpus、default和queries。corpus配置具有文本id和图片特征，default配置具有查询id、文本id和分数特征，而queries配置具有查询id和查询文本特征。每个配置都有训练集 split，且提供了相应的数据文件路径。

The dataset consists of three configurations: corpus, default, and queries. The corpus configuration has text-id and image features, the default configuration has query-id, text-id, and score features, and the queries configuration has query-id and query text features. Each configuration includes a training set split and provides the corresponding data file paths.