README.md

---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: image
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: generated_captions_1
    dtype: string
  - name: generated_captions_2
    dtype: string
  - name: comparison_1
    dtype: int64
  - name: comparison_2
    dtype: int64
  splits:
  - name: train
    num_bytes: 1178796
    num_examples: 12
  download_size: 835389
  dataset_size: 1178796
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---