metadata
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 1830471473
num_examples: 14657
- name: test
num_bytes: 378914937.6652264
num_examples: 2932
download_size: 2204019346
dataset_size: 2209386410.6652265
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*