vidore/shiftproject_test_beir
收藏Hugging Face2025-02-12 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/vidore/shiftproject_test_beir
下载链接
链接失效反馈官方服务：
资源简介：
这个数据集是BEIR版本的vidore/shiftproject_test数据集。它包含三个部分：corpus、qrels和queries。corpus部分包含文本和图像数据，qrels部分包含查询ID、语料库ID和分数，用于评估信息检索系统，queries部分包含查询ID和查询文本。在测试分割中，corpus部分有999个样本，qrels部分有100个样本，queries部分也有100个样本。
This dataset is the BEIR version of the vidore/shiftproject_test dataset. It consists of three parts: corpus, qrels, and queries. The corpus part contains text and image data, the qrels part contains query IDs, corpus IDs, and scores for evaluating information retrieval systems, and the queries part contains query IDs and query text. In the test split, the corpus part has 999 samples, the qrels part has 100 samples, and the queries part also has 100 samples.
提供机构：
vidore
