---
dataset_info:
  features:
  - name: caption
    dtype: string
  - name: coco_image
    dtype: image
  - name: sd3_image
    dtype: image
  - name: sd21_image
    dtype: image
  - name: sdxl_image
    dtype: image
  - name: dalle_image
    dtype: image
  - name: midjourney_image
    dtype: image
  splits:
  - name: train
    num_bytes: 3967077819.0
    num_examples: 7000
  download_size: 3963922174
  dataset_size: 3967077819.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---