README.md

---
dataset_info:
  features:
  - name: custom_id
    dtype: string
  - name: question
    dtype: string
  - name: question_images
    sequence: image
  - name: question_categories
    sequence: string
  - name: final_answer
    dtype: string
  - name: rationale_text
    dtype: string
  - name: rationale_images
    sequence: image
  - name: image_attribution
    dtype: string
  - name: subquestions_1_text
    dtype: string
  - name: subquestions_1_answer
    dtype: string
  - name: subquestions_2_text
    dtype: string
  - name: subquestions_2_answer
    dtype: string
  - name: subquestions_3_text
    dtype: string
  - name: subquestions_3_answer
    dtype: string
  - name: subquestions_4_text
    dtype: string
  - name: subquestions_4_answer
    dtype: string
  - name: subquestions_5_text
    dtype: string
  - name: subquestions_5_answer
    dtype: string
  - name: author_name
    dtype: string
  - name: author_email_address
    dtype: string
  - name: author_institution
    dtype: string
  splits:
  - name: train
    num_bytes: 77282884.0
    num_examples: 100
  download_size: 74705643
  dataset_size: 77282884.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---