metadata
dataset_info:
features:
- name: document_id
dtype: string
- name: document_text
dtype: string
- name: document_filename
dtype: string
- name: document_metadata
struct:
- name: file_size
dtype: int64
splits:
- name: train
num_bytes: 62271583
num_examples: 4997
download_size: 35900637
dataset_size: 62271583
configs:
- config_name: default
data_files:
- split: train
path: data/train-*