README.md

---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: split_name
    dtype: string
  - name: dataset_name
    dtype: string
  - name: base_question
    dtype: string
  - name: base_ans
    dtype: string
  - name: final_question
    dtype: string
  - name: final_ans
    dtype: string
  - name: difficulty_our
    dtype: string
  - name: difficulty_base
    dtype: string
  splits:
  - name: train
    num_bytes: 1096931
    num_examples: 21
  download_size: 856976
  dataset_size: 1096931
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---