metadata
dataset_info:
features:
- name: cls
dtype: int64
- name: image
dtype: image
splits:
- name: validation
num_bytes: 1148768134
num_examples: 50000
- name: train
num_bytes: 25457554354.125
num_examples: 1281167
download_size: 26631136693
dataset_size: 26606322488.125
configs:
- config_name: default
data_files:
- split: validation
path: data/validation-*
- split: train
path: data/train-*