metadata
dataset_info:
features:
- name: name
dtype: string
- name: image
dtype: image
- name: annotation
dtype: image
- name: question
dtype: string
- name: bbox
dtype: string
splits:
- name: train
num_bytes: 16589205
num_examples: 288
- name: validation
num_bytes: 4204902
num_examples: 73
download_size: 20780209
dataset_size: 20794107
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*