---
dataset_info:
  features:
  - name: source
    dtype: image
  - name: target
    dtype: image
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 146981825.0
    num_examples: 931
  download_size: 134011680
  dataset_size: 146981825.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---