metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 87079032.628
num_examples: 10004
download_size: 87515082
dataset_size: 87079032.628
configs:
- config_name: default
data_files:
- split: train
path: data/train-*