metadata
dataset_info:
features:
- name: image_id
dtype: int64
- name: words
sequence: string
- name: bboxes
sequence:
sequence: int64
- name: labels
sequence: float64
- name: image
dtype: binary
- name: image_path
dtype: string
- name: source
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 419370110
num_examples: 528
- name: test
num_bytes: 110356586
num_examples: 133
download_size: 509864102
dataset_size: 529726696