Datasets:

shodiBoy
/

diffusion.studio.dataset

license: mit
task_categories:
  - image-generation
language:
  - en
size_categories:
  - n<1K
dataset_info:
  features:
    - name: image
      dtype: image
    - name: text
      dtype: string
    - name: model
      dtype: string
    - name: category
      dtype: string
    - name: sub_category
      dtype: string
  splits:
    - name: train
      num_bytes: 18397027
      num_examples: 16
  download_size: 18293528
  dataset_size: 18397027
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Diffusion Studio Generations

This dataset contains 16 high-quality AI-generated images created using various diffusion models. Each image is accompanied by detailed prompts and model information, showcasing different styles and techniques in AI image generation.

Dataset Description

Curated by: shodiBoy
Language: English
License: MIT
Size: 16 images
Task: Image Generation
Repository: diffusion_studio_generations

Dataset Structure

Each entry in the dataset contains:

id: Unique identifier for the entry
prompt: Detailed text prompt used to generate the image
image: Path to the generated image
model: AI model used to generate the image
category: Main category of the image (Sports, Fashion, Cinematic)
sub_category: More specific classification

Models Used

flux-1-1-pro: Black Forest Labs FLUX.1 Pro model
GPT-IMAGE-1: OpenAI's GPT Image Generation model
RECRAFT-V3: Recraft V3 model

Image Categories

The dataset includes various types of images:

Sports: Alpine/skiing scenes, basketball action shots
Fashion: Portraits, avant-garde fashion photography
Cinematic: Sci-fi scenes, urban photography, vintage aesthetics

Usage

This dataset can be used for:

Training image generation models
Prompt engineering research
Model comparison studies
Computer vision research
Style analysis and categorization

File Structure

dataset/
├── images/
│   └── chosen/      # Generated images (16 files)
├── metadata/        # Individual JSON files for each image
├── dataset.json     # Main dataset file with all entries
├── dataset_info.json # Dataset metadata
└── README.md        # This file

Citation

If you use this dataset in your research, please cite:

@dataset{diffusion_studio_generations_2024,
  author = {shodiBoy},
  title = {Diffusion Studio Generations},
  year = {2024},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/shodiBoy/diffusion_studio_generations}
}