antonioloison/filtered-colpali-with-translations

该数据集包含两部分配置：corpus和queries。corpus部分包含图像和原始文本特征，以及图像的语言描述。数据集分为训练集，共有38686个示例，大小为15347018179.82字节。queries部分包含正例和负例文本段落、原始文本、原始语言、原始查询、翻译语言和翻译查询。该部分数据集同样分为训练集和测试集，训练集有109233个示例，测试集有500个示例。

The dataset consists of two configurations: corpus and queries. The corpus part includes image and original text features, as well as the language description of the image. The dataset is split into a training set with 38686 examples, totaling 15347018179.82 bytes in size. The queries part includes positive and negative text passages, original text, original language, original query, translation languages, and translated queries. This part of the dataset is also split into a training set with 109233 examples and a test set with 500 examples.