README.md

metadata

license: mit
task_categories:
  - question-answering
dataset_info:
  features:
    - name: qid
      dtype: int64
    - name: url
      dtype: string
    - name: title
      dtype: string
    - name: creationTime
      dtype: int64
    - name: followerCount
      dtype: int64
    - name: viewCount
      dtype: int64
    - name: numAnswers
      dtype: int64
    - name: numMachineAnswers
      dtype: int64
    - name: isLocked
      dtype: bool
    - name: isTrendyQuestion
      dtype: bool
    - name: asker
      struct:
        - name: uid
          dtype: int64
        - name: givenName
          dtype: string
        - name: familyName
          dtype: string
        - name: isMachineAnswerBot
          dtype: bool
    - name: answers
      dtype: string
  splits:
    - name: train
      num_bytes: 45900481870
      num_examples: 2971156
  download_size: 16052225826
  dataset_size: 45900481870
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

QUORA_ONE_MANY_QA

This dataset is derived from quora.com questioning data. It is a question with multiple answers.

The project provide gas for mnbvc.

STATISTICS

Raw data size

• 1000000 16G
• 2000000 17G
• 3000000 15G
• Updating...