---
dataset_info:
  features:
  - name: question_id
    dtype: string
  - name: category
    dtype: string
  - name: episode_history
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: extra_answers
    list: string
  - name: videos
    list: image
  splits:
  - name: train
    num_bytes: 8323898424
    num_examples: 1636
  download_size: 8314766007
  dataset_size: 8323898424
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---