metadata
dataset_info:
features:
- name: image
dtype: image
- name: caption
dtype: string
splits:
- name: train
num_bytes: 6732811
num_examples: 101
download_size: 6714241
dataset_size: 6732811
configs:
- config_name: default
data_files:
- split: train
path: data/train-*