metadata
dataset_info:
features:
- name: guid
sequence: string
- name: words
sequence: string
- name: labels
sequence: int64
- name: boxes
sequence:
sequence: int64
- name: actual_bboxes
sequence:
sequence: int64
- name: file_name
dtype: string
- name: image
sequence:
sequence:
sequence: uint8
splits:
- name: train
num_bytes: 98185080
num_examples: 594
- name: val
num_bytes: 5264337
num_examples: 32
- name: test
num_bytes: 57252300
num_examples: 347
download_size: 67438968
dataset_size: 160701717