---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: condition
    dtype: image
  - name: caption
    dtype: string
  splits:
  - name: train
    num_bytes: 21464858262.0
    num_examples: 40000
  - name: val
    num_bytes: 1055324771.0
    num_examples: 2000
  download_size: 22618735426
  dataset_size: 22520183033.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
---