metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: line_id
dtype: int64
- name: text
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 1390136247.188
num_examples: 4041
download_size: 1446691924
dataset_size: 1390136247.188
configs:
- config_name: default
data_files:
- split: train
path: data/train-*