README.md

---
dataset_info:
  features:
  - name: question_id
    dtype: string
  - name: video_path
    dtype: string
  - name: area
    dtype: string
  - name: conversations
    list:
    - name: role
      dtype: string
    - name: value
      dtype: string
  splits:
  - name: train
    num_bytes: 3009379
    num_examples: 7392
  - name: test
    num_bytes: 7765932
    num_examples: 19140
  download_size: 792760
  dataset_size: 10775311
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---