---
dataset_info:
  features:
  - name: souce_img
    dtype: image
  - name: edited_img
    dtype: image
  - name: edit_prompt
    dtype: string
  - name: pq_score
    dtype: int32
  - name: sc_score
    sequence: float32
  - name: o_score
    dtype: float32
  splits:
  - name: train
    num_bytes: 264746654.0
    num_examples: 100
  download_size: 264761615
  dataset_size: 264746654.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---