metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 1772796352.33
num_examples: 15803
download_size: 1773009285
dataset_size: 1772796352.33
configs:
- config_name: default
data_files:
- split: train
path: data/train-*