metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 111634441.5
num_examples: 17444
- name: test
num_bytes: 24043300
num_examples: 3752
download_size: 26252389
dataset_size: 135677741.5
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*