---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: annotation
    dtype: image
  splits:
  - name: train
    num_bytes: 31921472686.2
    num_examples: 16902
  - name: validation
    num_bytes: 3518885330.856
    num_examples: 1902
  download_size: 34143093932
  dataset_size: 35440358017.056
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---