metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 12827098747.644
num_examples: 9686
download_size: 12990218990
dataset_size: 12827098747.644
configs:
- config_name: default
data_files:
- split: train
path: data/train-*