mteb/JinaVDRGitHubReadmeRetrieval

这是一个多语言数据集，包含了多种语言的文本和图像数据。每个语言的数据集都分为三个部分：corpus（语料库），qrels（查询相关度），和queries（查询）。语料库部分包含了图像和对应的ID，而查询相关度部分包含了查询ID、语料库ID和相关度分数。查询部分包含了查询的ID和文本。数据集的测试分割包含了每个部分的示例数量和大小。数据集支持多种语言，包括阿拉伯语、孟加拉语、德语、英语、西班牙语、法语、印地语、印度尼西亚语、意大利语、日语、韩语、荷兰语、葡萄牙语、俄语、泰语、越南语和中文。

This is a multilingual dataset that includes text and image data in various languages. Each language dataset is divided into three parts: corpus, qrels, and queries. The corpus part includes images and corresponding IDs, while the qrels part includes query IDs, corpus IDs, and relevance scores. The queries part includes query IDs and text. The test splits of the dataset include the number of examples and size for each part. The dataset supports multiple languages, including Arabic, Bengali, German, English, Spanish, French, Hindi, Indonesian, Italian, Japanese, Korean, Dutch, Portuguese, Russian, Thai, Vietnamese, and Chinese.