---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: ground_truth
    dtype: string
  splits:
  - name: train
    num_bytes: 134344674.54817215
    num_examples: 3183
  - name: validation
    num_bytes: 33767206.01482785
    num_examples: 796
  download_size: 157959389
  dataset_size: 168111880.563
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---