README.md

metadata

dataset_info:
  features:
    - name: image
      dtype: image
  splits:
    - name: train
      num_bytes: 3054750999.892
      num_examples: 19406
  download_size: 4825600812
  dataset_size: 3054750999.892
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*