[R-sig-eco] documentation available that further explains error messages?

Ben Bolker bolker at zoology.ufl.edu
Wed May 7 23:20:20 CEST 2008


   A couple more suggestions:

  Mark Bravington's "debug" package (mvbdebug, available
on CRAN) can be a more powerful debugging tool if you
want to do a more graphical stepping-through your code;

   a cruder but effective technique is to put a cat()
statement into your objective function, e.g.

  objfun <- function(params,...) {
   ## calculate neg log likelihood
   neg.log.likel <- ...
   cat(params[1],params[2],neg.log.likel,"\n")
   neg.log.likel
}

   that way you can see where the function is getting in
trouble.



Kingsford Jones wrote:
> Hi Jeremy,
> 
> Since R is open source, the code is usually the best reference when
> you can't decipher an error message.  Often the easiest way to see
> what is going on is to type traceback() after receiving the error
> message -- this will show you the calls that led to the error.
> 
> Another very useful tool is to type
> debug(name.of.function.to.be.debugged) prior to calling the function
> of interest.
> The next time you use that function, debug will take you step-by-step
> through the function calls while allowing full interactive ability so
> you can, e.g. look at the structure of objects as they are created
> within the function (using, e.g., the str function).
> 
> Another option is to open the function's source file in a text window
> and use Find in your text editor to find the error.
> 
> Finally, searching for the error message in the help archives can
> often be fruitful.  That technique just led me to this:
> 
> https://stat.ethz.ch/pipermail/r-help/2006-January/086747.html
> 
> which indicates what the '[2]' is referring to.
> 
> 
> Kingsford Jones
> 
> 
> On Wed, May 7, 2008 at 12:30 PM, Jeremy Claisse <jclaisse at yahoo.com> wrote:
>> I'm trying to figure out how to overcome this error message:
>>
>>  Error in optim(c(5, -0.5), llfunc.prop.female.age, method = "BFGS") :
>>   non-finite finite-difference value [2]
>>
>>  And particularly would like to know what the "[2]" is pointing towards, I'm
>>  assuming that will help me search out the problem.  Is there any
>>  documentation available that explains the particulars of error messages?
>>
>>  Thanks,
>>  Jeremy
>>
>>  Jeremy Claisse
>>  Hawaii Cooperative Fishery Research Unit
>>  University of Hawaii
>>   <mailto:claisse at hawaii.edu> claisse at hawaii.edu
>>
>>
>>
>>         [[alternative HTML version deleted]]
>>
>>  _______________________________________________
>>  R-sig-ecology mailing list
>>  R-sig-ecology at r-project.org
>>  https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>>
> 
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: <https://stat.ethz.ch/pipermail/r-sig-ecology/attachments/20080507/aef9ad29/attachment.bin>


More information about the R-sig-ecology mailing list