metadata
dataset_info:
features:
- name: key
dtype: string
- name: zenodo_record_id
dtype: string
- name: zenodo_filename
dtype: string
- name: page_number
dtype: int64
- name: text
dtype: int64
- name: visual
dtype: int64
- name: mixed
dtype: int64
splits:
- name: train
num_bytes: 18856
num_examples: 179
download_size: 6775
dataset_size: 18856
configs:
- config_name: default
data_files:
- split: train
path: data/train-*