metadata
dataset_info:
features:
- name: image
dtype: image
- name: mask
dtype: image
splits:
- name: train
num_bytes: 3773157787.242
num_examples: 10798
- name: test
num_bytes: 2584899
num_examples: 20
download_size: 3784979907
dataset_size: 3775742686.242
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*