metadata
dataset_info:
features:
- name: id
dtype: int32
- name: image
dtype: image
- name: caption
dtype: string
- name: caption_ls
list: string
- name: category
dtype: string
- name: en_caption
dtype: string
splits:
- name: train
num_bytes: 1714820964399.125
num_examples: 7463023
download_size: 1708517206932
dataset_size: 1714820964399.125
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
copied from QuoQA-NLP/KoCC12M