README.md

metadata

dataset_info:
  features:
    - name: title
      dtype: string
    - name: question_id
      dtype: int64
    - name: question_body
      dtype: string
    - name: question_score
      dtype: int64
    - name: question_date
      dtype: string
    - name: answer_id
      dtype: int64
    - name: answer_body
      dtype: string
    - name: answer_score
      dtype: int64
    - name: answer_date
      dtype: string
    - name: tags
      sequence: string
  splits:
    - name: train
      num_bytes: 2142466142
      num_examples: 987122
  download_size: 829547986
  dataset_size: 2142466142
task_categories:
  - question-answering
language:
  - en
size_categories:
  - 100K<n<1M

Dataset Card for "stackoverflow_python"

Dataset Summary

This dataset comes originally from kaggle. It was originally split into three tables (CSV files) (Questions, Answers, and Tags) now merged into a single table. Each row corresponds to a pair (question-answer) and their associated tags.

The dataset contains all questions asked between August 2, 2008 and Ocotober 19, 2016.

Supported Tasks and Leaderboards

This might be useful for open-domain question-answering tasks.

Additional information

License

All Stack Overflow user contributions are licensed under CC-BY-SA 3.0 with attribution required.