---
dataset_info:
  features:
  - name: img_id
    dtype: string
  - name: object_img
    dtype: binary
  - name: target_img
    dtype: binary
  - name: original_img
    dtype: binary
  splits:
  - name: train
    num_bytes: 5939149109
    num_examples: 61796
  download_size: 5931946047
  dataset_size: 5939149109
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---