metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: int64
splits:
- name: test
num_bytes: 131280000
num_examples: 40000
- name: train
num_bytes: 101323112
num_examples: 240000
download_size: 129737247
dataset_size: 232603112
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*