---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': object_image
          '1': original_image
          '2': target_image
  splits:
  - name: train
    num_bytes: 795680313.664
    num_examples: 127312
  download_size: 4071257308
  dataset_size: 795680313.664
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---