metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 1610276662.625
num_examples: 17125
- name: test
num_bytes: 1741142678.19
num_examples: 16135
download_size: 3747351407
dataset_size: 3351419340.815
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*