metadata
dataset_info:
features:
- name: image
dtype: image
splits:
- name: train
num_bytes: 20567854200
num_examples: 52000
- name: test
num_bytes: 6099994350
num_examples: 14000
download_size: 16252108052
dataset_size: 26667848550
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*