metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 993704177.182
num_examples: 32626
- name: test
num_bytes: 14925431
num_examples: 500
download_size: 1026467780
dataset_size: 1008629608.182
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*