---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: question_type
    dtype: string
  - name: visual_indices
    sequence: int64
  - name: images
    sequence: image
  splits:
  - name: train
    num_bytes: 91356720.0
    num_examples: 400
  download_size: 77787520
  dataset_size: 91356720.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---