metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 22143947000.998
num_examples: 28393
download_size: 6707338199
dataset_size: 22143947000.998
configs:
- config_name: default
data_files:
- split: train
path: data/train-*