README.md

metadata

dataset_info:
  features:
    - name: Prompt
      dtype: string
    - name: Image
      dtype: image
    - name: Upsampled_Prompt
      dtype: string
    - name: Image_With_Upsampled_Prompt
      dtype: image
    - name: model_name
      dtype: string
    - name: seed
      dtype: int64
  splits:
    - name: train
      num_bytes: 625589974
      num_examples: 200
  download_size: 625589110
  dataset_size: 625589974
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Dataset Card for "drawbench-sdxl"

The dataset was generated using https://github.com/sayakpaul/caption-upsampling. Refer to the repository for more details.