metadata
dataset_info:
features:
- name: image
dtype: image
- name: text_caption
dtype: string
- name: uid
dtype: string
splits:
- name: train
num_bytes: 745575783
num_examples: 5400
- name: validation
num_bytes: 82841754
num_examples: 600
download_size: 827418656
dataset_size: 828417537
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*