[Basta-users] Errors when attempting to run multiple chains

[Colchero, Fernando]

Dear Jenni,

   Well, looking closely at the error, this is something we haven't had before. My first thought is that this has to do with the starting parameters; when you run more than one simulation, it sets initial parameters from different points, and maybe what's happening is that some of the initial parameters are way off. We were sure that this wouldn't happen anymore, but maybe your dataset has a different structure from what we had tried before. Try setting the argument "thetaStart" to c(-6, 0.001), or something along those values and let us know what happens. If this doesn't work, what we could do is that you could send us the dataset so we try to reproduce and fix the error. Of course, as soon as we fix the error we would erase the dataset from our computers. 

   Once again thanks for trying BaSTA and for posting the error! Best,

   Fernando


______________________________________________
 
Fernando Colchero

Max Planck Institute for Demographic Research
Konrad Zuse Str. 1, 18057, Rostock, Germany
Tel:  +49 (0) 381 2081 113
Fax: +49 (0) 381 2081 413
E-mail: Colchero at demogr.mpg.de
______________________________________________





On Mar 30, 2012, at 4:23 PM, Graham, Jennifer wrote:


	Dear Fernando,
	
	Thank you for your help.
	
	These are the package versions I used: snowfall_1.84 snow_0.3-8,  BaSTA_1.2,     msm_1.1
	and R version 2.14.2
	
	Regards
	
	Jenni
	________________________________________
	From: Colchero, Fernando [Colchero at demogr.mpg.de]
	Sent: 30 March 2012 14:20
	To: Graham, Jennifer
	Cc: basta-users at lists.r-forge.r-project.org
	Subject: Re: [Basta-users] Errors when attempting to run multiple chains
	
	Dear Jenni,
	
	  Thanks a lot for posting the error in the BaSTA mailing list. What version of BaSTA are you using? We had a bug in version 1.1 that usually gave that error. Would you mind installing version 1.2 and running it again? Please let us know if you get the error again; we can help you go through your data and find out where the problem might be.
	
	  Best,
	
	  Fernando
	
	______________________________________________
	
	Fernando Colchero
	
	Max Planck Institute for Demographic Research
	Konrad Zuse Str. 1, 18057, Rostock, Germany
	Tel:  +49 (0) 381 2081 113
	Fax: +49 (0) 381 2081 413
	E-mail: Colchero at demogr.mpg.de<mailto:Colchero at demogr.mpg.de>
	______________________________________________
	
	
	
	
	
	On Mar 30, 2012, at 2:24 PM, Graham, Jennifer wrote:
	
	Hi,
	
	I have a long-term mark-recapture dataset and the data checking function works great and I have been able to run single simulations. However when using the same data, and general model settings I get an error when I attempt to run multiple chains, I tried running with and without parallel= TRUE and I get a different error each time, see below for R Code used and errors.
	
	Thanks for your help and a great R package.
	
	Kind Regards
	
	Jenni
	
	newData<-DataCheck(DATA,studyStart=25,studyEnd=112,autofix=rep(1,7),silent=FALSE)
	
	Output from DataCheck:
	
	No problems were detected with the data.
	*DataSummary*
	- Number of individuals         =    1,460
	- Number with known birth year  =    1,351
	- Number with known death year  =       0
	- Number with known birth
	AND death years                =       0
	- Total number of detections
	in recapture matrix            =    7,697
	- Earliest detection time       =      25
	- Latest detection time         =     112
	- Earliest recorded birth year  =       1
	- Latest recorded birth year    =     109
	
	out<-basta(DATA,studyStart=25,studyEnd=112) #  This single simulation works
	
	out<-basta(DATA,studyStart=25,studyEnd=112,,nsim=4,parallel=TRUE,ncpus=4)#  I get the following error which occurs regardless  of the number of simulations I request or the ncpus value
	
	Error in checkForRemoteErrors(val) :
	2 nodes produced errors; first error: missing value where TRUE/FALSE needed
	
	#As I thought it could be due to the snowfall package I also ran parallel=FALSE and then get another error
	
	Error in if (post == -Inf) { : missing value where TRUE/FALSE needed
	
	
	
	--------------------------------------------------------------------------------
	Jennifer Graham
	PhD Researcher
	Centre for Ecology and Conservation
	University of Exeter, Cornwall Campus
	Tremough
	Penryn
	Cornwall
	TR10 9EZ
	_______________________________________________
	Basta-users mailing list
	Basta-users at lists.r-forge.r-project.org<mailto:Basta-users at lists.r-forge.r-project.org>
	https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/basta-users
	
	
	----------
	
	This mail has been sent through the MPI for Demographic Research. Should you receive a mail that is apparently from a MPI user without this text displayed, then the address has most likely been faked. If you are uncertain about the validity of this message, please check the mail header or ask your system administrator for assistance.
	
	
	
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/basta-users/attachments/20120330/4b4ed0dc/attachment-0001.html>