metadata
dataset_info:
features:
- name: id
dtype: int64
- name: words
sequence: string
- name: bboxes
sequence:
sequence: float64
- name: ner_tags
sequence: int64
- name: image
dtype: image
splits:
- name: train
num_bytes: 1288055514.25
num_examples: 1222
- name: test
num_bytes: 53446678
num_examples: 50
download_size: 1321443890
dataset_size: 1341502192.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*