Which Cuda Do I NEED for license plate?? Inconsistent Documentation

[dannyeverest]

Windows 10 Pro:

I downloaded cpai 2.9.5 and the license plate module gavre me this error on install:

23:41:25:ALPR: W1224 23:41:25.808676 9348 dynamic_loader.cc:285] Note: [Recommend] copy cudnn into CUDA installation directory.
23:41:25:ALPR: For instance, download cudnn-10.0-windows10-x64-v7.6.5.32.zip from NVIDIA's official website,
23:41:25:ALPR: then, unzip it and copy it into C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0
23:41:25:ALPR: You should do this according to your CUDA installation directory and CUDNN version.
23:41:25:ALPR: Traceback (most recent call last):
23:41:25:ALPR: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 105, in
23:41:25:ALPR: ALPR_adapter().start_loop()
23:41:25:ALPR: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\codeproject_ai_sdk\module_runner.py", line 278, in start_loop
23:41:25:ALPR: self.initialise()
23:41:25:ALPR: File "C:\Program Files\CodeProject\AI\modules\ALPR\ALPR_adapter.py", line 33, in initialise
23:41:25:ALPR: if paddle.device.get_cudnn_version() / 100.0 < self.opts.min_cuDNN_version:
23:41:25:ALPR: File "C:\Program Files\CodeProject\AI\modules\ALPR\bin\windows\python39\venv\lib\site-packages\paddle\device_init_.py", line 179, in get_cudnn_version
23:41:25:ALPR: cudnn_version = int(core.cudnn_version())
23:41:25:ALPR: RuntimeError: (PreconditionNotMet) The third-party dynamic library (cudnn64_8.dll) that Paddle depends on is not configured correctly. (error code is 126)
23:41:25:ALPR: Suggestions:
23:41:25:ALPR: 1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
23:41:25:ALPR: 2. Configure third-party dynamic library environment variables as follows:
23:41:25:ALPR: - Linux: set LD_LIBRARY_PATH by export LD_LIBRARY_PATH=...
23:41:25:ALPR: - Windows: set PATH by `set PATH=XXX; (at ..\paddle\phi\backends\dynload\dynamic_loader.cc:312)
23:41:35:ALPR_adapter.py: W1224 23:41:35.121275 2364 dynamic_loader.cc:285] Note: [Recommend] copy cudnn into CUDA installation directory.
23:41:35:ALPR_adapter.py: For instance, download cudnn-10.0-windows10-x64-v7.6.5.32.zip from NVIDIA's official website,
23:41:35:ALPR_adapter.py: then, unzip it and copy it into C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0
23:41:35:ALPR_adapter.py: You should do this according to your CUDA installation directory and CUDNN version.

On the main CPai it says to use the latest cuda, on another page it says we need some 11.* version.... and the error is telling me I need to add a file to my cuda 10 directory? so I need cuda 10 for license plate reader????

I just want this to work. From windows to docker dockerdesktop, to linux docker now back to windows.... Pls someone help.

I must be missing something.

[dannyeverest]

I followed the instructions from the error message, and it didn't work. does anyone know where to find the documentation for license plate reader?