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: 4517752044.335
num_examples: 4041
download_size: 257645534
dataset_size: 4517752044.335
configs:
- config_name: default
data_files:
- split: train
path: data/train-*