metadata
dataset_info:
features:
- name: video_index
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 22774
num_examples: 193
download_size: 3057
dataset_size: 22774
configs:
- config_name: default
data_files:
- split: train
path: data/train-*