metadata
dataset_info:
features:
- name: title
dtype: string
- name: description
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 729639660
num_examples: 5226
download_size: 679019809
dataset_size: 729639660
configs:
- config_name: default
data_files:
- split: train
path: data/train-*