---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 358607914.596
    num_examples: 1404
  - name: validation
    num_bytes: 101516965.0
    num_examples: 396
  - name: test
    num_bytes: 51295610.0
    num_examples: 200
  download_size: 513446998
  dataset_size: 511420489.596
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---