metadata
dataset_info:
features:
- name: question
dtype: string
- name: options
sequence: string
- name: correct_answer
dtype: string
- name: image
dtype: image
- name: description
dtype: string
- name: image_file_name
dtype: string
- name: textbook_name
dtype: string
- name: original_image_file_name
dtype: string
- name: shuffled_image_file_name
dtype: string
splits:
- name: train
num_bytes: 1544363
num_examples: 12
download_size: 1546221
dataset_size: 1544363
configs:
- config_name: default
data_files:
- split: train
path: data/train-*