clevrer / README.md
zechen-nlp's picture
Upload dataset
6ad1fb7 verified
|
raw
history blame
1.79 kB
metadata
dataset_info:
  - config_name: counterfactual
    features:
      - name: video
        dtype: string
      - name: question_id
        dtype: string
      - name: question_type
        dtype: string
      - name: question
        dtype: string
      - name: program
        sequence: string
      - name: choices
        sequence:
          - name: choice_id
            dtype: int64
          - name: choice
            dtype: string
          - name: program
            sequence: string
          - name: answer
            dtype: string
      - name: conversations
        sequence:
          - name: from
            dtype: string
          - name: value
            dtype: string
    splits:
      - name: train
        num_bytes: 29562437
        num_examples: 18642
      - name: validation
        num_bytes: 14840172
        num_examples: 9333
    download_size: 4019296
    dataset_size: 44402609
  - config_name: predictive
    features:
      - name: video
        dtype: string
      - name: question_id
        dtype: string
      - name: question_type
        dtype: string
      - name: question
        dtype: string
      - name: program
        sequence: string
      - name: choices
        sequence:
          - name: choice_id
            dtype: int64
          - name: choice
            dtype: string
          - name: program
            sequence: string
          - name: answer
            dtype: string
      - name: conversations
        sequence:
          - name: from
            dtype: string
          - name: value
            dtype: string
    splits:
      - name: train
        num_bytes: 6940854
        num_examples: 7179
      - name: validation
        num_bytes: 3457628
        num_examples: 3557
    download_size: 924628
    dataset_size: 10398482
configs:
  - config_name: counterfactual
    data_files:
      - split: train
        path: counterfactual/train-*
      - split: validation
        path: counterfactual/validation-*
  - config_name: predictive
    data_files:
      - split: train
        path: predictive/train-*
      - split: validation
        path: predictive/validation-*