---
dataset_info:
  features:
  - name: source
    dtype: image
  - name: target
    dtype: image
  - name: mask
    dtype: image
  - name: ghost
    dtype: image
  - name: concatenated
    dtype: image
  splits:
  - name: train
    num_bytes: 2910803165.0
    num_examples: 998
  - name: test
    num_bytes: 36278282.0
    num_examples: 14
  download_size: 2927940401
  dataset_size: 2947081447.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---