Running the MDNet demo on CPU

[nitish11]

Hi,

How to run the model on CPU?

I tried your instruction
You can also disable the GPU support by setting opts.useGpu in 'tracking/mdnet_init.m' to false (not recommended).

But, still I am getting error in running demo_tracking

Trial>> demo_tracking
Initialization...
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Warning: While loading an object of class 'gpuArray':
There is a problem with the CUDA driver or with this GPU device. Be sure that you have a supported GPU and that the latest driver
is installed. 
> In mdnet_init (line 68)
  In mdnet_run (line 29)
  In demo_tracking (line 22) 
Error using gpuArray/gather
Attempting to access the property or method of an invalid object.

Error in vl_simplenn_move>@(x)gather(x) (line 17)
  case 'cpu', moveop = @(x) gather(x) ;

Error in vl_simplenn_move (line 26)
          net.layers{l}.(f) = moveop(net.layers{l}.(f)) ;

Error in mdnet_init (line 89)
    net_conv = vl_simplenn_move(net_conv, 'cpu') ;

Error in mdnet_run (line 29)
[net_conv, net_fc, opts] = mdnet_init(img, net);

Error in demo_tracking (line 22)
result = mdnet_run(conf.imgList, conf.gt(1,:), net);

[jzhoubu]

Hi,I use MATLAB 2016a and VS 2015 C++ ,and have switched "opts.useGpu" to false.
But I got the same error as @nitish11 did.

[hyseob]

@nitish11 @sysu-zjw
Hi,
Have you tried removing "vl_simplenn_move" lines? (line 89-90 in mdnet_init)
The model parameters are stored in the matlab format of "gpuArray", and I did not tested it on the environment where no gpu is installed.
If you have a gpu installed, you may need to check the CUDA version.

[jzhoubu]

@HyeonseobNam
I have tried your advice. However, it didn't work. I think it may relate to the version of matlab.
I will try the GPU version. Thanks for sharing! :)

[salmanmaq]

A straight-forward solution would be if you can provide us the model .mat files in the CPU format i.e. they contain normal MatLab arrays, not gpuArrays. That would resolve this issue.

Computers without GPUs cannot read gpuArrays, and would throw an error when they encounter it, as they do in the case of .mat files.

[hyseob]

@nitish11 @sysu-zjw @salmanmaq
For now you can download the model files in the CPU format from:
https://drive.google.com/open?id=0B4G5mwcHYoSuMXo2eUpkbkpTWGc
Thanks :)

[salmanmaq]

Thank you so much. Just what I have been waiting for. :)
I'll test it out and update and comment.

[SoulmateB]

@salmanmaq
Hi,have you got the the model .mat files in the CPU format? I have the same problem when I run the demo on CPU, could you tell me how to solve it ? Thank you !!

[SoulmateB]

@HyeonseobNam
How to run the model on CPU? I have the same problem when I run the demo on CPU.....Thank you !!

[salmanmaq]

@SoulmateB I did this a few months back. The files HyeonseobNam provided in the Drive worked fine for me. Did you try them out?

[SoulmateB]

No,I can't download the files,could you provide me the files? Thank you !!

[salmanmaq]

I just checked and the link works for me. Do you encounter any particular error with the link?

[afunTW]

I downloaded the model in CPU format and applied it in demo_tracking.m
My Parallel Computing Toolbox version is 6.9 in Matlab R2016b

However, when I run the demo_tracking.m, the program will idle over 30 minutes and leads the parallel pool close. Am I in the right way to run this program in CPU mode?