---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 4340493430.975
    num_examples: 9995
  - name: validation
    num_bytes: 1808061.0
    num_examples: 5
  download_size: 3469197006
  dataset_size: 4342301491.975
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---