metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: int32
splits:
- name: train
num_bytes: 154806589362.012
num_examples: 1281167
download_size: 146527689212
dataset_size: 154806589362.012
configs:
- config_name: default
data_files:
- split: train
path: data/train-*