metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: caption
dtype: string
- name: code
dtype: string
- name: image
dtype: image
- name: uri
dtype: string
- name: origin
dtype: string
- name: date
dtype: timestamp[us]
splits:
- name: train
num_bytes: 690355633
num_examples: 48296
download_size: 620253367
dataset_size: 690355633