metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 5863021.833333333
num_examples: 500
- name: val
num_bytes: 1168940.1666666667
num_examples: 100
download_size: 6125119
dataset_size: 7031962