dlquantizer workflow failed when running example code

I have tried running the "Quantize a Neural Network" example from this link -- https://www.mathworks.com/help/deeplearning/ref/dlquantizer.html, but I receive the following error when running the validate function:

Error using dlquantizer/validate
dlquantizer workflow failed.

I have tried a few other example codes and no matter what it is failing in the same place. I am unable to pause in the debugger as the error seems to occur while the function is running dos commands.

KFA

• .m file containing the script I am running with the complete example, failing on line 67
• .mat file containing the trained network and both IMDS (though these are generated by the script)
• .txt file containing output from the 'ver' command

@Asvin Kumar here is the information you requested re: https://www.mathworks.com/matlabcentral/answers/770448-error-dlquantizer-workflow-failed-deep-network-quantizer-app

[Dor Rubin]

Hi Will,

Can you report the output of the script below? My guess is that there is an issue with the gpu setup like a missing dependency or CUDA compute capability of 6.1 or 6.3 and higher (https://www.mathworks.com/help/deeplearning/ug/quantization-workflow-prerequisites.html)

>> gpuDevice
>> gpuEnvObj = coder.gpuEnvConfig;
>> gpuEnvObj.GpuId = 1;
>> gpuEnvObj.BasicCodegen = 1;
>> gpuEnvObj.BasicCodeexec = 1;
>> results = coder.checkGpuInstall(gpuEnvObj)

[Will Kaczmarek]

ans = 

  CUDADevice with properties:

                      Name: 'TITAN RTX'
                     Index: 1
         ComputeCapability: '7.5'
            SupportsDouble: 1
             DriverVersion: 11.2000
            ToolkitVersion: 11
        MaxThreadsPerBlock: 1024
          MaxShmemPerBlock: 49152
        MaxThreadBlockSize: [1024 1024 64]
               MaxGridSize: [2.1475e+09 65535 65535]
                 SIMDWidth: 32
               TotalMemory: 2.5770e+10
           AvailableMemory: 1.8744e+10
       MultiprocessorCount: 72
              ClockRateKHz: 1770000
               ComputeMode: 'Default'
      GPUOverlapsTransfers: 1
    KernelExecutionTimeout: 1
          CanMapHostMemory: 1
           DeviceSupported: 1
           DeviceAvailable: 1
            DeviceSelected: 1

Compatible GPU           : PASSED 
CUDA Environment         : PASSED 
	Runtime   : PASSED 
	cuFFT     : PASSED 
	cuSOLVER  : PASSED 
	cuBLAS    : PASSED 

[Will Kaczmarek]

@Dor Rubin @Asvin Kumar Any ideas?

Answer 1

[Madhav Thakker]

Hi Will,

I understand you are facing error when running the validate function. The developers are aware about this bug and will try to fix it in future releases.

Hope this helps.

[Will Kaczmarek]

Thank you, we were able to get it working with some help from Matlab Support. I appreciate you taking the time to answer!