---
dataset_info:
  features:
  - name: original_image
    dtype: image
  - name: conditioning_image
    dtype: image
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 27276930122.696
    num_examples: 24826
  download_size: 24475639540
  dataset_size: 27276930122.696
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---