---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 395071392.6819575
    num_examples: 9965
  - name: validation
    num_bytes: 43474688.981042534
    num_examples: 1108
  download_size: 426472219
  dataset_size: 438546081.663
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---