---
dataset_info:
  features:
  - name: tar_img
    dtype: image
  - name: edit_instruction
    dtype: string
  - name: src_img
    dtype: image
  splits:
  - name: train
    num_bytes: 48363557552.22
    num_examples: 152210
  download_size: 48578385934
  dataset_size: 48363557552.22
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---