metadata
dataset_info:
features:
- name: image
dtype: image
- name: mask
dtype: image
splits:
- name: train
num_bytes: 91362033.72656922
num_examples: 930
- name: valid
num_bytes: 22846921.273430783
num_examples: 233
download_size: 113415827
dataset_size: 114208955
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*