---
dataset_info:
  features:
  - name: source_dataset
    dtype: string
  - name: input_image
    dtype: image
  - name: edit_instruction
    dtype: string
  - name: output_reasoning_text
    dtype: string
  - name: output_image
    dtype: image
  splits:
  - name: train
    num_bytes: 1240986042.0
    num_examples: 2000
  download_size: 1177492608
  dataset_size: 1240986042.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---