metadata
dataset_info:
features:
- name: image_url
dtype: string
- name: caption
dtype: string
splits:
- name: train
num_bytes: 2030
num_examples: 10
download_size: 4034
dataset_size: 2030
configs:
- config_name: default
data_files:
- split: train
path: data/train-*