metadata
language:
- en
dataset_info:
features:
- name: id
dtype: int64
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 671914857
num_examples: 10000
download_size: 882416418
dataset_size: 671914857
configs:
- config_name: default
data_files:
- split: train
path: data/train-*