---
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-*
---