metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: string
splits:
- name: train
num_bytes: 16158122
num_examples: 60000
download_size: 15320084
dataset_size: 16158122
configs:
- config_name: default
data_files:
- split: train
path: data/train-*