README.md

metadata

license: mit
task_categories:
  - summarization
  - text2text-generation
  - text-generation
  - text-classification
language:
  - en
pretty_name: OnionNewsScrape

This is a dataset of Onion news articles:

Note

• The headers and body of the news article is split by a ' #~# ' token
• Lines with just the token had no body or no header and can be skipped
• Feel free to use the script provided to scape the latest version, it takes about 30 mins on an i7-6850K