README.md

metadata

dataset_info:
  features:
    - name: id
      dtype: string
    - name: question
      dtype: string
    - name: choices
      dtype: string
    - name: correct_answer
      dtype: int64
    - name: image
      dtype: image
  splits:
    - name: train
      num_bytes: 130688805359.58
      num_examples: 678034
    - name: test
      num_bytes: 1290885818.416
      num_examples: 6676
  download_size: 106501046765
  dataset_size: 131979691177.996
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*