Getting error for NVIDIA CudNN with Matlab 2021b in Windows 10, NIVIDIA T500

27 views (last 30 days)
Christian Schwarz
Christian Schwarz on 7 Apr 2022
Answered: Sayan Saha on 12 May 2022
Similar to described in the previous post, I installed all MATLAB apps, 3rd party software and checked all environemental variables but I am unable to passe the Deep-learning code generation check (cuDNN and tensorrt).
I would realle appreciate your help?
Below, you can find the list of errors:
1. cuDNN
gpuEnvObj = coder.gpuEnvConfig;
gpuEnvObj.GpuId = 0;
gpuEnvObj.DeepLibTarget = 'cudnn';
gpuEnvObj.DeepCodegen = 1;
gpuEnvObj.DeepCodeexec = 1;
coder.checkGpuInstall(gpuEnvObj)
Compatible GPU : PASSED
CUDA Environment : PASSED
Runtime : PASSED
cuFFT : PASSED
cuSOLVER : PASSED
cuBLAS : PASSED
cuDNN Environment : PASSED (Warning: Deep learning code generation has been tested with cuDNN v8.1. The provided cuDNN library v8.4 may not be fully compatible.)
Deep Learning (cuDNN) Code Generation: FAILED (Test GPU code generation failed with the error 'emlc:compilationError'. View report for further information: View report)
2. tensorrt
envCfg.DeepLibTarget = 'tensorrt';
envCfg.DeepCodegen = 1;
envCfg.Quiet = 1;
coder.checkGpuInstall(envCfg);
Basic Code Generation : PASSED
Basic Code Execution : PASSED
ans =
struct with fields:
gpu: 1
cuda: 1
cudnn: 0
tensorrt: 0
basiccodegen: 1
basiccodeexec: 1
deepcodegen: 0
deepcodeexec: 0
tensorrtdatatype: 0
profiling: 0
Error using coder.checkGpuInstall (line 33)
One or more of the system checks did not pass, with the following errors ...
Deep Learning (TensorRT) Code Generation: (Test GPU code generation failed with the error 'emlc:compilationError'. View
report for further information: View report)
Error in GPUtest (line 16)
coder.checkGpuInstall(envCfg);
>> gpuDeviceTable
ans =
1×5 table
Index Name ComputeCapability DeviceAvailable DeviceSelected
_____ _____________ _________________ _______________ ______________
1 "NVIDIA T500" "7.5" true true
I am using an NVIDIA T500 and MS Visual Studio 2019 for the C-compiler

Answers (1)

Sayan Saha
Sayan Saha on 12 May 2022
Hi Christian,
The errors are likely coming from the version mis-match of the CuDNN library. As mentioned in the warning, we have only verified deep learning code-generation against the CuDNN library v8.1 for MATLAB R2021b. Can you try installing v8.1 version and check if you can proceed with code-generation?
Thanks,
Sayan

Products


Release

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by