---
dataset_info:
  features:
  - name: ghost_images
    dtype: image
  - name: target
    dtype: image
  - name: binary_mask
    dtype: image
  - name: prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 13721864.0
    num_examples: 12
  download_size: 13719348
  dataset_size: 13721864.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---