Drexel dragonThe Math ForumDonate to the Math Forum



Search All of the Math Forum:

Views expressed in these public forums are not endorsed by Drexel University or The Math Forum.


Math Forum » Discussions » Software » comp.soft-sys.matlab

Topic: Training/classification of data
Replies: 3   Last Post: May 10, 2013 9:08 AM

Advanced Search

Back to Topic List Back to Topic List Jump to Tree View Jump to Tree View   Messages: [ Previous | Next ]
Ilya Narsky

Posts: 133
Registered: 11/7/08
Re: Training/classification of data
Posted: May 10, 2013 9:08 AM
  Click to see the message monospaced in plain text Plain Text   Click to reply to this topic Reply

"Srey " <indu.kpk@gmail.com> wrote in message
news:kmhr11$ij8$1@newscl01ah.mathworks.com...
> Error using ==> svmtrain at 453
> Error calculating the kernel function:
> Out of memory. Type HELP MEMORY for your options.
>
> Error in ==> create_train_test_data svmStruct = svmtrain(data,class);
>


First, please type

which svmtrain

to verify that you are using svmtrain from the official MATLAB installation.

You don't say what release you are using. In older MATLAB releases, the
default solver for svmtrain was QP. This solver creates a full kernel matrix
of size N-by-N for N observations. Since you have so many observations, you
would run out of memory at this step.

Use the SMO solver instead. This is the default choice in recent MATLAB
versions. The 'kernelcachelimit' parameter for SMO lets you control the size
of the cached kernel matrix. If you set this parameter to a small value,
svmtrain may take a long time to finish. If you set it to a large value, you
may run out of memory. -Ilya






Point your RSS reader here for a feed of the latest messages in this topic.

[Privacy Policy] [Terms of Use]

© Drexel University 1994-2014. All Rights Reserved.
The Math Forum is a research and educational enterprise of the Drexel University School of Education.