metadata
dataset_info:
features:
- name: image
dtype: string
- name: mask
dtype: image
splits:
- name: train
num_bytes: 95998005.624
num_examples: 11416
- name: validation
num_bytes: 17605309.356
num_examples: 2006
download_size: 61225613
dataset_size: 113603314.97999999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*