nthakur/msmarco-passage-sampled-10k

--- language: - en license: cc-by-sa-3.0 task_categories: - text-retrieval source_datasets: - Tevatron/msmarco-passage --- # nthakur/msmarco-passage-sampled-10k This is a 10k randomly sampled training pairs of the Tevatron [msmarco-passage](https://huggingface.co/datasets/Tevatron/msmarco-passage) for debugging and training models on a smaller subset of MSMARCO training data. ## Citing & Authors Have a look at [Tevatron](https://github.com/texttron/tevatron). <!--- Describe where people can find more information -->

原始信息汇总

nthakur/msmarco-passage-sampled-10k

概述

• 语言: 英语
• 许可: cc-by-sa-3.0
• 任务类别: 文本检索
• 源数据集: Tevatron/msmarco-passage

描述

这是一个从Tevatron msmarco-passage中随机抽样的10,000个训练对的数据集，用于调试和在小规模子集上训练MSMARCO训练数据模型。