metadata
dataset_info:
features:
- name: image
dtype: image
- name: mask
dtype: image
splits:
- name: train
num_bytes: 209492691.08437952
num_examples: 2234
- name: valid
num_bytes: 52752151.41962048
num_examples: 559
download_size: 260917522
dataset_size: 262244842.504
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*