metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: string
splits:
- name: train
num_bytes: 39679327205
num_examples: 108754
download_size: 39508520000
dataset_size: 39679327205
configs:
- config_name: default
data_files:
- split: train
path: data/train-*