deu05232/msmarco-w-instructions-shuffle

这是一个包含查询和对应正负例文档的数据集，用于训练模型进行相关任务。数据集包含字段如查询ID、查询文本、正例文档信息（包括文档ID、解释、分数、联合ID、文本和标题）、负例文档信息（包括文档ID、文本和标题）、是否仅包含指令、是否仅包含查询以及是否包含指令的标记。数据集分为训练集，共有约980250个示例，大小为约12.22GB。

This is a dataset containing queries and corresponding positive and negative example documents for training models on related tasks. The dataset includes fields such as query ID, query text, positive document information (including document ID, explanation, score, joint ID, text, and title), negative document information (including document ID, text, and title), whether it contains only instructions, whether it contains only the query, and a marker indicating whether it contains instructions. The dataset is split into a training set with approximately 980,250 examples, totaling about 12.22GB in size.