README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 796698977.373
      num_examples: 123653
  download_size: 728294151
  dataset_size: 796698977.373
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*