---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: original_image
    dtype: binary
  - name: edited_image
    dtype: binary
  - name: initial_edit_mask
    dtype: binary
  - name: final_edit_mask
    dtype: binary
  - name: edit_instruction
    dtype: string
  splits:
  - name: train
    num_bytes: 2218764
    num_examples: 3
  download_size: 2244434
  dataset_size: 2218764
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---