YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Code for the Paper titled "Perpetual Misogyny: How Gendered Tropes Shape Text-To-Image AI Personalization"
Repository Structure
CIVITAI_VISUALIZATIONS/
βββ .virtual_documents/ # Temporary files from Jupyter
βββ data/ # Final curated datasets
β βββ subset1/ # Specific data splits or versions
β βββ subset2/
β βββ ...
βββ misc/
β βββ credentials/ # API keys and sensitive config (excluded from versioning)
βββ plots/ # Output plots and figures used in the paper
βββ βββ jupyter_notebooks/ # Main analysis notebooks
β βββ 0_Scraping_image_metadata.ipynb # Scrapes CivitAI metadata via API
β βββ Section_1_Figure_1_image_grid.ipynb # Grid of sample images
β βββ Section_3-2-1_Figure_3_histogram.ipynb # Histogram of upload trends
β βββ Section_3-2-1_Figure_4_Mivolo.ipynb # Model activity plot (Mivolo-focused)
β βββ Section_3-3-1_Figure_5_tags.ipynb # Tag frequency and usage visualizations
β βββ Section_3-3-3_download_popular_models.ipynb # Download models for analysis
β βββ Section_3-3-3_Figure_6.ipynb # Promotional tag usage patterns
β βββ Section_3-3-4_Figure_8a.ipynb # Ranking of popular models
β βββ Section_3-3-4_Figure_8b.ipynb # Continuation of model rankings
β βββ Section_3-3-4_Figure_9_Sankey.ipynb # Sankey diagram: user-model contributions
β βββ Section_3-3-4_LLM_annotation.ipynb # Annotations using large language models
β βββ Section_3-4_extract_LoRA_metadata.ipynb # LoRA metadata extraction
β βββ SuppM_Figure_12_Danbooru_Taxonomy.ipynb # Danbooru tag taxonomy: visualization
β βββ SuppM_Figure_13_Danbooru_taxonomy.ipynb # Tag grouping and structure
β βββ SuppM_Figure_13.ipynb # Supplementary figure generation
βββ misc/ # Utility scripts and API credentials (excluded)
βββ plots/ # Output plots and visualizations
βββ public/ # Optional public-facing files
βββ .gitignore
βββ .gitmodules
βββ README.md # This file
βββ requirements.txt # Python dependencies
Project Setup
Requirements
This project requires Python 3.8 or higher. Ensure you have it installed before proceeding.
Installation
Clone the Repository
git clone https://gitlab.uzh.ch/latent-canon/pm-paper.git cd pm-paperCreate a Virtual Environment
(Recommended to avoid dependency conflicts)python -m venv venv source venv/bin/activate # On macOS/Linux venv\Scripts\activate # On WindowsInstall Dependencies
pip install -r requirements.txtJupyter Notebook Setup (Optional)
If running Jupyter notebooks, ensure the environment is linked:python -m ipykernel install --user --name=venv --display-name "Python (venv)"
Notes
- Ensure you have necessary system dependencies installed (e.g.,
opencvmay require additional system libraries). - If you encounter any issues, ensure you're using the correct Python environment (
venvactivated).
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
π
Ask for provider support