presencesw/decompose_question_BAMBOOGLE_FERMI

该数据集名为decompose_question_BAMBOOGLE_FERMI，主要用于处理复杂问题的分解。每个样本包含一个复杂问题、一个实体列表、多个三元组（每个三元组包含一个问题）和一个答案。数据集的训练集包含8878个样本，总大小为1784299字节。

This dataset is named decompose_question_BAMBOOGLE_FERMI, which is primarily designed for complex question decomposition. Each sample comprises a complex question, an entity list, multiple triples (each containing a question), and an answer. The training split of this dataset contains 8878 samples with a total size of 1784299 bytes.

原始信息汇总

数据集概述

数据集信息

• 特征列表:

  • entities: 序列类型，值为 null
  • triplets: 列表类型，包含以下字段:
    • question: 字符串类型
  • answer: 字符串类型
  • complex_question: 字符串类型

• 数据分割:

  • train:
    • 字节数: 1784299
    • 样本数量: 8878

• 数据大小:

  • 下载大小: 344213 字节
  • 数据集大小: 1784299 字节

• 配置:

  • default:
    • 数据文件:
      • train: 路径为 data/train-*

数据样本结构

json { "complex_question": "Does Mercury help detect coronavirus?", "entities": [], "triples": [ { "question": "What is the name of the coronavirus?", }, { "question": "Does Mercury help detect COVID-19?", }, { "question": "What is mercury used to detect?", }, { "question": "What are some symtoms of coronavirus?", } ], "answer": "" }