metadata
dataset_info:
features:
- name: street
dtype: string
- name: tokens
sequence: string
- name: ner_tags
sequence:
class_label:
names:
'0': O
'1': B-STRASSE
'2': I-STRASSE
'3': B-HAUSNUMMER
'4': B-BUCHSTABE
'5': I-BUCHSTABE
'6': B-ZUSATZ
'7': I-ZUSATZ
splits:
- name: train
num_bytes: 209043716
num_examples: 1759075
download_size: 56492927
dataset_size: 209043716
configs:
- config_name: default
data_files:
- split: train
path: data/train-*