metadata
dataset_info:
features:
- name: cate
dtype: int64
- name: video
dtype: string
- name: video_type
dtype: float64
- name: question
dtype: string
- name: candidates
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 511379
num_examples: 1769
download_size: 38312
dataset_size: 511379
configs:
- config_name: default
data_files:
- split: train
path: data/train-*