QGIS installation paddy GPU error

Error content

Use the following code to check whether the paddle GPU is installed successfully

import paddle
paddle.utils.run_check()

Get the following error reports

RuntimeError: (PreconditionNotMet) The third-party dynamic library (cublas64_102.dll;cublas64_10.dll) that Paddle depends on is not configured correctly. (error code is 126)
  Suggestions:
  1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
  2. Configure third-party dynamic library environment variables as follows:
  - Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...`
  - Windows: set PATH by `set PATH=XXX; (at C:\home\workspace\Paddle_release\paddle\fluid\platform\dynload\dynamic_loader.cc:265)

It is found that other CONDA environments can be used normally, CUDA installation and environment variable configuration are correct, and the environment variable is checked in QGIS. It is found that there is no C: \program files\NVIDIA GPU computing toolkit\CUDA\v10.2\bin in path.

Solution:

For the desktop, you can add C:\program files\NVIDIA GPU computing toolkit\CUDA\v10.2\bin
after the path in QGIS\bin\qgis-ltr-bin.env. For plug-ins developed in vscode, you can add o4w in the same folder, Change set path to set path =% osgeo4w in env.bat_ROOT%\bin;% WINDIR%\system32;% WINDIR%;% WINDIR%\system32\WBem;% CUDA_BIN_PATH%