roop教程

install python (and pip too if neeed)
install git
install ffmpeg
If you are on Windows, install “Microsoft Visual C++ 14” and C++ build tools.
git clone https://github.com/s0md3v/roop
cd roop && pip install -r requirements.txt
Download this file and keep it in roop directory. Mirror #1, Mirror #2, Mirror #3, Mirror #4. Rename it to inswapper_128.onnx if it isn’t already.

Done. You should be able to use this software with python run.py command.

1 NVIDIA

If you are on windows, this detailed explanation will most likely work for you. If you are on linux, well read on:

install CUDA toolkit from here
install CUDNN from here
pip uninstall onnxruntime onnxruntime-gpu
pip install torch torchvision torchaudio --force-reinstall --index-url https://download.pytorch.org/whl/cu118
pip install onnxruntime-gpu

安装问题

我是win10操作系统，python3.9.5

cuda_11.8.0_522.06_windows.exe

下载玩安装cuda_11.8.0。如果之前有安装旧版本，需要先卸载旧版本后再安装。

cudnn-windows-x86_64-8.9.1.23_cuda11-archive.zip

下载好之后，直接解压cuDNN压缩包，可以看到bin、include、lib目录，如下图：

  打开C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA，找到bin、include、lib目录，如下图：


  接下来将cuDNN压缩包内对应的bin、include、lib目录下的文件对应的复制到bin、include、lib目录下。

● 注意：是复制文件夹下的文件到bin、include、lib目录，不是复制目录。

进入 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8\extras\demo_suite

在此路径下打开powershell（shift+鼠标右键选择在此处打开Powershell窗口（S）），执行命令.\bandwidthTest.exe

怎么解决 Could not locate zlibwapi.dll. Please make sure it is in your library path!

Installing Zlib

Zlib is a data compression software library that is needed by cuDNN.

Download and extract the zlib package from here
If using Chrome, Edge, or other modern browsers, the file may not automatically download. If this happens, right-click the link and choose Save link as…. Then, paste the URL into a browser window.
Add the directory path of zlibwapi.dll to the environment variable PATH.
比如：C:\Windows