metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: image
dtype: image
- name: text
sequence: string
- name: file_id
dtype: string
splits:
- name: train
num_bytes: 715798376.168
num_examples: 7092
download_size: 693617401
dataset_size: 715798376.168