---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: solution
    struct:
    - name: text
      dtype: string
  - name: original_question
    dtype: string
  - name: original_answer
    dtype: string
  - name: question
    dtype: string
  - name: language
    dtype: string
  - name: source
    dtype: string
  splits:
  - name: train
    num_bytes: 2312116874.0
    num_examples: 917
  download_size: 2309363105
  dataset_size: 2312116874.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---