README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: frame_id
    dtype: string
  - name: video_id
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: question_type
    dtype: string
  splits:
  - name: train
    num_bytes: 251267471
    num_examples: 360
  - name: validation
    num_bytes: 31407904
    num_examples: 45
  - name: test
    num_bytes: 31408966
    num_examples: 45
  download_size: 314085572
  dataset_size: 314084341
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---