metadata
dataset_info:
features:
- name: id
dtype: int64
- name: words
sequence: string
- name: bboxes
sequence:
sequence: float64
- name: ner_tags
sequence:
class_label:
names:
'0': I-COMPANY
'1': I-DATE
'2': I-ADDRESS
'3': I-TOTAL
'4': O
- name: image
dtype: image
splits:
- name: train
num_bytes: 980745292
num_examples: 897
- name: test
num_bytes: 53446678
num_examples: 50
download_size: 1019860538
dataset_size: 1034191970
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*