Comments for Parallelcoding.com

Comment on Genetic Algorithms on GPU using CUDA by janisozaur

janisozaur — Wed, 22 Feb 2012 14:21:22 +0000

the link is not working, is there a chance of fixing this?

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by lily

lily — Sat, 18 Feb 2012 16:41:00 +0000

It's ok. . Currently the programmer is not member of their team. I will try to solve the problem.

I apologise for bothering you. If I found the solution, I will report the result here for others that may have a similar problem.

Comment on Particle Swarm Optimization on the GPU using CUDA by Robert Green

Robert Green — Sat, 18 Feb 2012 13:14:19 +0000

It seems to me that there is only a small chunk of pseudo-code in that paper. If you share your code, I'd be happy to help.

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by Robert Green

Robert Green — Sat, 18 Feb 2012 13:13:06 +0000

Lily,

I'm sorry you've been unable to get this to work. I would contact the creator of cuSVM as well.

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by lily

lily — Sat, 18 Feb 2012 10:03:31 +0000

After such a long stressing effort I failed to use cuSVM in 64-bit MATLAB.

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by lily

lily — Sat, 18 Feb 2012 09:58:53 +0000

I think that output arguments of the mexFunction is allocated properly, for example for plhs[0] we have:

plhs[0]=mxCreateNumericMatrix(numSVs, 1,mxSINGLE_CLASS, mxREAL);

float *SvAlphas=(float*)mxGetData(plhs[0]);

in this link ( http://www.mathworks.com/matlabcentral/newsreader/view_thread/166716 ) I found a similar problem with "mxCreateNumericMatrix" but I don't know what is the correct solution.

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by lily

lily — Sat, 18 Feb 2012 09:49:24 +0000

Hi,

I test the produced *.mexw64 file, with a sample input training data. unfortunately this error is appeared, and I don't know what is going wrong.

??? One or more output arguments not assigned during call to "cusvm_cuda_mex".

The test code that I wrote is:

load heart_scale.mat;

data = heart_scale_inst;
value = heart_scale_label;

C = 10;
kernel = 0.5;
stopcrit = 0.001;

[alphas, beta, svs] = cusvm_cuda_mex(value, data, C, kernel, [], stopcrit);

Comment on Particle Swarm Optimization on the GPU using CUDA by yasaman

yasaman — Fri, 17 Feb 2012 19:44:25 +0000

have you tried the code prepared in reference 8, "Evaluation of parallel particle swarm optimization algorithms within the CUDA(TM) architecture" ?
I cannot build the code, could you help me please?

Comment on Installing the MATLAB plug-in for CUDA in Ubuntu 10.04 by lily

lily — Fri, 10 Feb 2012 21:25:13 +0000

I found that cuSVM has not any cross-validation function, but the libsvm has this feature.

Should I implement the cross validation function using the cuSVMTrain.mexw64 mex-function?

If so, what is the correct searching range of two parameters C and Lambda?

Unfortunately cuSVM documentation is too brief and did not mentions any cross-validation procedure.

Your help and comments are greatly appreciated.

Comment on cuSVM in Visual Studio 2010 with CUDA 4.0 by lily

lily — Fri, 10 Feb 2012 19:37:45 +0000

it's my pleasure.