metadata
dataset_info:
features:
- name: image
dtype: image
- name: label
dtype: image
splits:
- name: train
num_bytes: 21346977
num_examples: 254
download_size: 19917130
dataset_size: 21346977
configs:
- config_name: default
data_files:
- split: train
path: data/train-*