---
dataset_info:
  features:
  - name: id
    dtype: int32
  - name: image
    dtype: image
  - name: choices
    list: string
  - name: answer_type
    dtype: string
  - name: answers
    struct:
    - name: A
      dtype: float32
    - name: B
      dtype: float32
    - name: C
      dtype: float32
    - name: D
      dtype: float32
    - name: E
      dtype: float32
  - name: year
    dtype: int32
  splits:
  - name: train
    num_bytes: 212298417
    num_examples: 400
  download_size: 211832329
  dataset_size: 212298417
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---