metadata
dataset_info:
features:
- name: id
dtype: string
- name: image_name
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: image
dtype: image
- name: prompt
dtype: string
- name: answers
dtype: string
splits:
- name: train
num_bytes: 240550374
num_examples: 8850
- name: test
num_bytes: 58560718
num_examples: 2213
download_size: 205992821
dataset_size: 299111092
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*