---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 14064418868.420666
    num_examples: 534903
  - name: validation
    num_bytes: 5598819.7373351725
    num_examples: 201
  download_size: 13499025275
  dataset_size: 14070017688.158
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---