small Gradio web user interface used for generating images
Modify the line number 17 and edit the path where the model is saved
cache_dir='/media/agp/d58e0f56-1cd5-45af-938c-27e43b4fc343/kandinsky/tmp',
IE if on windows :
cache_dir='c:\kandinskymodels',
python 3.10.6 has to be installed. you can download an install from : https://www.python.org/downloads/release/python-3106/
Install steps :
- mkdir kandinsky && cd kandinsky
- python -m venv venv
- source venv/bin/activate
- git clone https://github.com/ai-forever/Kandinsky-2.git
- cd Kandinsky-2
- pip install "git+https://github.com/ai-forever/Kandinsky-2.git"
- pip install git+https://github.com/openai/CLIP.git
- pip install gradio
- wget https://raw.githubusercontent.com/adrianpuiu/kandinsky/main/app.py 10 python app.py
open and point your browser at : http://127.0.0.1:7860
Enjoy generating images
- pip install jupyterlab
- jupiter-lab
- open the browser and go to notebooks folder and select the one you want