metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: string
splits:
- name: train
num_bytes: 1251594891
num_examples: 20000
download_size: 1210167602
dataset_size: 1251594891
configs:
- config_name: default
data_files:
- split: train
path: data/train-*