---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: question_idx
    dtype: string
  - name: question
    dtype: string
  - name: images
    sequence: image
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 1513227715.047
    num_examples: 1921
  download_size: 1034321183
  dataset_size: 1513227715.047
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---