metadata
dataset_info:
features:
- name: qa_id
dtype: string
- name: image
dtype: image
- name: image_url
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: question_type
dtype: string
- name: category
dtype: string
- name: keyword
dtype: string
- name: annotator
dtype: string
- name: group
dtype: int64
- name: answer_label
dtype: string
- name: answer_text
dtype: string
- name: choices
dtype: string
- name: distractors
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 48294828
num_examples: 377
download_size: 48052893
dataset_size: 48294828
configs:
- config_name: default
data_files:
- split: train
path: data/train-*