metadata
dataset_info:
features:
- name: name
dtype: string
- name: state_name
dtype: string
- name: state_abbreviation
dtype: string
- name: postal_code
dtype: string
- name: city
dtype: string
- name: street_number
dtype: string
- name: street_name
dtype: string
- name: tokens
sequence: string
- name: ner_tags
sequence: string
- name: ocr_text
dtype: string
- name: type
dtype: string
- name: unit_number
dtype: string
- name: unit_designator
dtype: string
splits:
- name: train
num_bytes: 286066554
num_examples: 720000
- name: eval
num_bytes: 31785288
num_examples: 80000
- name: test
num_bytes: 79442813
num_examples: 200000
download_size: 156775401
dataset_size: 397294655
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
- split: test
path: data/test-*