metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: image
dtype: image
- name: idx
dtype: int64
- name: text
dtype: string
splits:
- name: train
num_bytes: 21219899
num_examples: 18
download_size: 0
dataset_size: 21219899