metadata
dataset_info:
features:
- name: image
dtype: image
- name: captions
dtype: string
splits:
- name: train
num_bytes: 139293745.3069307
num_examples: 12609
download_size: 137375063
dataset_size: 139293745.3069307
configs:
- config_name: default
data_files:
- split: train
path: data/train-*