joshuapsa/gpt-generated-news-paragraphs-v1.1

--- configs: - config_name: default data_files: - split: train path: data/train-* - split: valid path: data/valid-* - split: test path: data/test-* dataset_info: features: - name: class_name dtype: string - name: text dtype: string - name: aviation dtype: class_label: names: '0': '0' '1': '1' - name: cybersecurity dtype: class_label: names: '0': '0' '1': '1' - name: domestic_unrest_violence dtype: class_label: names: '0': '0' '1': '1' - name: extreme_weather dtype: class_label: names: '0': '0' '1': '1' - name: forced_labor dtype: class_label: names: '0': '0' '1': '1' - name: general_biz_trend dtype: class_label: names: '0': '0' '1': '1' - name: individual_accidents_tragedies dtype: class_label: names: '0': '0' '1': '1' - name: later_report dtype: class_label: names: '0': '0' '1': '1' - name: lawsuit_legal_insurance dtype: class_label: names: '0': '0' '1': '1' - name: leisure_other_news dtype: class_label: names: '0': '0' '1': '1' - name: maritime dtype: class_label: names: '0': '0' '1': '1' - name: pandemics_large_scale_diseases dtype: class_label: names: '0': '0' '1': '1' - name: railway dtype: class_label: names: '0': '0' '1': '1' - name: strike dtype: class_label: names: '0': '0' '1': '1' - name: trade_war_embargos_bans dtype: class_label: names: '0': '0' '1': '1' - name: transportation_trends_projects dtype: class_label: names: '0': '0' '1': '1' - name: war_conflict dtype: class_label: names: '0': '0' '1': '1' - name: warehouse_fire dtype: class_label: names: '0': '0' '1': '1' - name: class_index dtype: class_label: names: '0': '0' '1': '1' - name: label sequence: int64 splits: - name: train num_bytes: 419816 num_examples: 720 - name: valid num_bytes: 52468 num_examples: 90 - name: test num_bytes: 52223 num_examples: 90 download_size: 179362 dataset_size: 524507 --- # Dataset Card for "gpt-generated-news-paragraphs-v1.1" - This dataset was created solely for the purpose of code testing. - This dataset was generated from prompting chatGPT to create sample pieces of news setences according to a topic. - Sample prompt: "generate 50 paragraphs on the topic of "very recent breaking news on wars and conflicts events" with some sample location names. One example: "a missile struck near a residential building in Kiev last night. Russia denied Ukraine's accusations of attacking non-military targets"" - The output paragraphs were then used to construct huggingface dataset. - Changes from v1.0: added column `class_name` for ease of use in downstream tasks [More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

原始信息汇总

数据集概述

数据集配置

• 默认配置：
  • 训练集：路径为 data/train-*
  • 验证集：路径为 data/valid-*
  • 测试集：路径为 data/test-*

数据集信息

• 特征：

  • class_name：类别名称，数据类型为字符串
  • text：文本内容，数据类型为字符串
  • aviation：航空，数据类型为类别标签，取值为 0 或 1
  • cybersecurity：网络安全，数据类型为类别标签，取值为 0 或 1
  • domestic_unrest_violence：国内动荡暴力，数据类型为类别标签，取值为 0 或 1
  • extreme_weather：极端天气，数据类型为类别标签，取值为 0 或 1
  • forced_labor：强迫劳动，数据类型为类别标签，取值为 0 或 1
  • general_biz_trend：一般商业趋势，数据类型为类别标签，取值为 0 或 1
  • individual_accidents_tragedies：个人事故悲剧，数据类型为类别标签，取值为 0 或 1
  • later_report：后续报道，数据类型为类别标签，取值为 0 或 1
  • lawsuit_legal_insurance：诉讼法律保险，数据类型为类别标签，取值为 0 或 1
  • leisure_other_news：休闲其他新闻，数据类型为类别标签，取值为 0 或 1
  • maritime：海事，数据类型为类别标签，取值为 0 或 1
  • pandemics_large_scale_diseases：大流行病大规模疾病，数据类型为类别标签，取值为 0 或 1
  • railway：铁路，数据类型为类别标签，取值为 0 或 1
  • strike：罢工，数据类型为类别标签，取值为 0 或 1
  • trade_war_embargos_bans：贸易战禁运禁令，数据类型为类别标签，取值为 0 或 1
  • transportation_trends_projects：交通趋势项目，数据类型为类别标签，取值为 0 或 1
  • war_conflict：战争冲突，数据类型为类别标签，取值为 0 或 1
  • warehouse_fire：仓库火灾，数据类型为类别标签，取值为 0 或 1
  • class_index：类别索引，数据类型为类别标签，取值为 0 或 1
  • label：标签，数据类型为整数

• 数据集划分：

  • 训练集：字节数为 419816，样本数为 720
  • 验证集：字节数为 52468，样本数为 90
  • 测试集：字节数为 52223，样本数为 90

• 数据集大小：

  • 下载大小：179362 字节
  • 数据集大小：524507 字节