---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 158341201.0
    num_examples: 320
  - name: val
    num_bytes: 39693554.0
    num_examples: 81
  download_size: 197965214
  dataset_size: 198034755.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---