Aesthetic Shadow is a 1.1b parameters visual transformer designed to evaluate the quality of anime images. It accepts high-resolution 1024x1024 images as input and provides a prediction score that quantifies the aesthetic appeal of the artwork. Leveraging cutting-edge deep learning techniques, this model excels at discerning fine details, proportions, and overall visual coherence in anime illustrations. Base url https://huggingface.co/shadowlilac/aesthetic-shadow
Installing requires Python 3.10. You can download Python here https://www.python.org/. Then follow these steps:
- Clone the repository
git clone https://github.com/gluttony-10/aesthetic-shadow.git
- Navigate into the cloned directory
cd aesthetic-shadow
- Set up a virtual environment
python -m venv venv
- Activate the new venv:
- Windows:
venv\scripts\activate
- Unix based systems:
source venv/bin/activate
- Windows:
- Install the requirements
pip install -r requirements.txt
- Pull changes
git pull
- Activate the venv
venv\scripts\activate
- Re-Install the requirements
pip install -r requirements.txt
- Activate the venv
venv\scripts\activate
- For single image
python inference.py --image text/test.jpg
- For batch
python batch.py --path test --output outputs --threshold 0.5 --batch 8