metadata
dataset_info:
features:
- name: text
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 15799603
num_examples: 15
download_size: 15798678
dataset_size: 15799603
configs:
- config_name: default
data_files:
- split: train
path: data/train-*