---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: mask
    dtype: image
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 2904778554.228332
    num_examples: 18422
  - name: test
    num_bytes: 2583231.0
    num_examples: 20
  download_size: 2678439361
  dataset_size: 2907361785.228332
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---