README.md

metadata

annotations_creators:
  - crowdsourced
language:
  - ru
language_creators:
  - crowdsourced
license:
  - cc0-1.0
multilinguality:
  - monolingual
pretty_name: Habr QnA
size_categories:
  - 100K<n<1M
source_datasets:
  - original
tags: []
task_categories:
  - text-generation
  - question-answering
task_ids:
  - language-modeling
  - open-domain-qa

Dataset Card for Habr QnA

Table of Contents

• Dataset Card for Habr QnA
  • Table of Contents
  • Dataset Description
    • Dataset Summary
    • Languages
  • Dataset Structure
    • Data Fields
    • Data Splits
  • Dataset Creation
  • Additional Information
    • Dataset Curators

Dataset Description

• Repository: https://github.com/its5Q/habr-qna-parser

Dataset Summary

This is a dataset of questions and answers scraped from Habr QnA. There are 723430 asked questions with answers, comments and other metadata.

Languages

The dataset is mostly Russian with source code in different languages.

Dataset Structure

Data Fields

Data fields can be previewed on the dataset card page.

Data Splits

All 723430 examples are in the train split, there is no validation split.

Dataset Creation

The data was scraped with a script, located in my GitHub repository

Additional Information

Dataset Curators

• https://github.com/its5Q