GUIDE FOR CURRENT VERSION (CUDA ON NVIDIA)

[MostlyWhat]

Just a quick info thingy for anyone reading for the setup that worked for me.

DISCLAIMER: THIS IS FOR RUNNING CUDA ON NVIDIA

Environment: Use venv like the tutorial is fine or anaconda will work too

System Dependencies: Deep-Live-Cam = Latest Clone from GitHub NVIDIA CUDA = 11.8 (Use Offline Installer) NVIDIA CUDNN = 8.5.0.96 (Extract Zip into the CUDA folder)

Python Dependencies (Fixes the face_enhancer missing):

Uninstall These (Run Them Anyways) pip uninstall basicsr -y pip uninstall gfpgan -y pip uninstall onnxruntime onnxruntime-gpu

Reinstall w/ Github pip install git+https://github.com/xinntao/BasicSR.git@master pip install git+https://github.com/TencentARC/GFPGAN.git@master pip install onnxruntime-gpu==1.16.3

Steps:

1. Do the system dependencies first (apart from repo)
2. Just Follow the tutorial

Suggestions: Write any issues down in the comments and I will try to update it.