metadata
dataset_info:
features:
- name: id
dtype: string
- name: videos
list: image
- name: question
dtype: string
- name: choices
list: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 6503945654
num_examples: 510
download_size: 6178737669
dataset_size: 6503945654
configs:
- config_name: default
data_files:
- split: train
path: data/*.parquet