metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 1524500887.48
num_examples: 174440
- name: test
num_bytes: 4012025
num_examples: 574
download_size: 1431305551
dataset_size: 1528512912.48
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*