metadata
dataset_info:
features:
- name: doc_type
dtype: string
- name: doc_id
dtype: string
- name: num
dtype: string
- name: date_str
dtype: string
- name: description
dtype: string
- name: url_metadata
dtype: string
- name: lang
dtype: string
- name: url_pdf
dtype: string
- name: doc_number
dtype: string
splits:
- name: train
num_bytes: 1015438
num_examples: 3934
download_size: 151811
dataset_size: 1015438
configs:
- config_name: default
data_files:
- split: train
path: data/train-*