---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: target
    dtype: string
  - name: images
    sequence: image
  splits:
  - name: train
    num_bytes: 4643307196.391
    num_examples: 31097
  download_size: 4658717628
  dataset_size: 4643307196.391
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---