metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 1529036386
num_examples: 174440
- name: test
num_bytes: 4026949
num_examples: 574
download_size: 1431639335
dataset_size: 1533063335
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*