---
dataset_info:
  features:
  - name: original_image
    dtype: image
  - name: edit_prompt
    dtype: string
  - name: new_image
    dtype: image
  splits:
  - name: train
    num_bytes: 15287553.0
    num_examples: 550
  - name: test
    num_bytes: 4336319.0
    num_examples: 151
  - name: valid
    num_bytes: 1661844.0
    num_examples: 54
  download_size: 21245949
  dataset_size: 21285716.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---