README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: solution
      dtype: string
  splits:
    - name: train
      num_bytes: 24246698
      num_examples: 6000
    - name: validation
      num_bytes: 8079367
      num_examples: 2000
    - name: test
      num_bytes: 8082295
      num_examples: 2000
  download_size: 35134652
  dataset_size: 40408360
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*
license: apache-2.0
task_categories:
  - visual-question-answering
size_categories:
  - 1K<n<10K