metadata
dataset_info:
features:
- name: id
dtype: string
- name: image
dtype: string
- name: question
dtype: string
- name: answers
list: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 826007
num_examples: 7424
- name: validation
num_bytes: 103243
num_examples: 928
- name: test
num_bytes: 103842
num_examples: 928
download_size: 165184
dataset_size: 1033092
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*