How to solve "Unexpected error calling cuDNN: CUDNN_STATUS_EXECUTION_FAILED." error?
    6 views (last 30 days)
  
       Show older comments
    
    santhosh kumar buddepu
 on 9 Aug 2023
  
    
    
    
    
    Commented: santhosh kumar buddepu
 on 13 Aug 2023
            Hello mathworks team,
Iam using NVIDIA RTX A4000 GPU and Matlab 2018b version
As part of learning I have taken the example of Transfer learning using alexnet from matlab documentation. I used the same example they have given but at training the network Iam getting an error.
netTransfer = trainNetwork(augimdsTrain,layers,options);
ERROR:Error using trainNetwork (line 150)
Unexpected error calling cuDNN: CUDNN_STATUS_EXECUTION_FAILED.
Error in alexnet_test_matlabexample (line 57)
netTransfer = trainNetwork(augimdsTrain,layers,options);
and, after I have given some more inputs provided in this below link
try
    nnet.internal.cnngpu.reluForward(1);
catch ME
end
parpool('local', gpuDeviceCount);
spmd
    try, nnet.internal.cnngpu.reluForward(1); catch ME, end
end
still I am getting same error. please help me in this regard.
0 Comments
Accepted Answer
  Joss Knight
    
 on 11 Aug 2023
        The Ada GPU architecture is not supported in R2018b. You need to upgrade MATLAB. You should have received a warning about this when you first started using the device.
More Answers (0)
See Also
Categories
				Find more on Deep Learning Toolbox in Help Center and File Exchange
			
	Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
