metadata
dataset_info:
features:
- name: images
dtype: image
- name: descriptions
dtype: string
splits:
- name: train
num_bytes: 2289043360
num_examples: 204
download_size: 131982879
dataset_size: 2289043360
configs:
- config_name: default
data_files:
- split: train
path: data/train-*