Skip to content

kernlab - error message: array(0, c(n, p)) : 'dim' specifies too large an array

6 messages · David Winsemius, Martin Batholdy, Steve Lianoglou

Original
Martin Batholdy
# 
Hi,

For another trainingset I get this error message, which again is rather cryptic to me:


 Setting default kernel parameters  

Error in array(0, c(n, p)) : 'dim' specifies too large an array
RMate stopped at line 0 of selection
Calls: rvm ... .local -> backsolve -> as.matrix -> chol -> diag -> array



thanks for any suggestions!
Martin Batholdy
# 
I am using a linear kernel (vanilladot).
By switching the kernel, I actually get rid of the error message, but I would like to stick to the linear one ...
On 13.02.2012, at 16:23, Martin Batholdy wrote:

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      David Winsemius
    

    

      
      #
    

  


  

      
On Feb 13, 2012, at 10:23 AM, Martin Batholdy wrote:

        

            

      
      
      
    

        
Just imagine how it seems to us!

            

      
      
      
    

        
You are on you way to the prize for the greatest number of cryptic  
(your word) postings in a short interval. (And this one doesn't even  
have the context of your posting of 8 minutes ago.)

            

      
      
      
    

        
More details about data and code.

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Martin Batholdy
    

    

      
      #
    

  


  

      
Ok, I am sorry,

My trainingset consists of a 60 x 204 matrix (independent_training ? 204 features).
I have 60 continuous labels (dependent_training, ranging from 2.25 to 135).

this is all the code I use:

library(kernlab)
rvm(as.matrix(independent_training), dependent_training, type="regression", kernel = "vanilladot")

        
On 13.02.2012, at 16:40, David Winsemius wrote:

        

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Steve Lianoglou
    

    

      
      #
    

  


  

      
Hi,

On Mon, Feb 13, 2012 at 10:53 AM, Martin Batholdy

        
<batholdy at googlemail.com> wrote:

            

      
      
      
    

        
Can you call `traceback()` after you get the error to see if you can
follow the code path that results in the explosion?

Downloading the kernlab src package will be helpful while your smoking
out the error so you can look at the entire source code, too.

In my .Rprofile, I actually have something like so:

options(error=utils:::dum.frames)

Which allows me to call `debugger()` after an error is thrown and
drops me into the location that threw the error (most of the time (I
think)), allowing me to poke around and see who's who, and what's
what.

HTH,
-steve

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Steve Lianoglou
    

    

      
      #
    

  


  

      
Sorry, this:

            

      
      
      
    

        
Should be:

options(error=utils:::dump.frames)

-steve