---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: 'Unnamed: 0'
    dtype: int64
  - name: similarity
    dtype: int64
  - name: expected_prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 372835157.7
    num_examples: 4010
  download_size: 355736993
  dataset_size: 372835157.7
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---