[ADMB Users] ADMB development cycle (bug in file f1b2vc1)

[Ben Bolker]

  I guess the point is that it would be nice to set things up so you
weren't *required* to be diligent and move things over explicitly, but
rather to mirror automatically or (ideally) use a single SVN repository.

  If it's (1) not practical to open the Honolulu SVN repository for
world-readability, (2) not safe to use google code as the official
repository, would it be possible to write scripts to mirror Honolulu to
Google code on a daily basis ... ?

 I'd also be curious to hear answers to Jon's questions (e.g., how could
one build ADMB from the sources?)

  cheers
    Ben Bolker

John Sibert wrote:
> Jon (et al),
> The explanation is simple. The code on code.google has not been kept up 
> to date with the "official" svn repository in Honolulu. We could be a 
> bit more diligent I suppose.
> 
> I generally eat my own dog food.
> John
> 
> Schnute, Jon wrote:
>> We've had some excellent exchanges among ADMB users recently. In
>> particular, according to a message from Dave on August 6 (attached at
>> the bottom of this message), Ben found a fatal bug. A few days later
>> (August 10), Johnoel announced the release of ADMB 9.0.363 with "various
>> fixes to the libraries". I made the assumption that this bug had
>> probably been fixed, so I thought I'd check the status of the code at
>> http://code.google.com/p/admb-project/. In particular, I looked at the
>> file f1b2vc1.cpp, located at:
>>
>> http://code.google.com/p/admb-project/source/browse/trunk/df1b2-separabl
>> e/f1b2vc1.cpp
>>
>> According to the SVN log on this file, it hasn't changed since release
>> 43 on March 31. Also, it appears not to include Dave's suggestion about
>> ADUNCONST(dvector,x).
>>
>> So my question is this: How does the source code site
>> http://code.google.com/p/admb-project/ fit into the ADMB development
>> cycle? Also, how would I actually build a version of ADMB from the files
>> in the trunk subdirectories: df1b2-separable, examples, LICENSE,
>> linad99, nh99, and tools99?
>>
>> I've used SVN to build and track my own local version of the source
>> files, but I see from the change log at:
>>
>> http://code.google.com/p/admb-project/source/list 
>>
>> that all changes since release 43 have pertained to the Wiki.
>>
>> Probably, I've failed to notice an explanation somewhere of how all this
>> should work, and I suspect that the Google code site is still in
>> development. But I know from my own recent experience that the tools
>> available at "Google code" can greatly facilitate project development.
>> Perhaps the ADMB user community would explore some of the changes Ben
>> discussed if they/we had a full understanding of the complete toolkit
>> needed to build ADMB from the latest source code.
>>
>> Thanks,
>> Jon
>>
>> ******************************************
>> Jon Schnute
>> Pacific Biological Station
>> 3190 Hammond Bay Road
>> Nanaimo, BC V9T 6N7
>> CANADA
>>
>> NEW email: Jon.Schnute at dfo-mpo.gc.ca 
>>  (formerly schnutej at pac.dfo-mpo.gc.ca)
>> ******************************************
>>
>> -----Original Message-----
>> From: users-bounces at admb-project.org
>> [mailto:users-bounces at admb-project.org] On Behalf Of dave fournier
>> Sent: August-06-09 8:30 AM
>> To: users at admb-project.org
>> Subject: [ADMB Users] bug in file f1b2vc1
>>
>>
>> Thanks Ben you found a fatal bug. Hard to believe that function had
>> never been used before. In the file f1b2vc1.cpp
>> the line
>>
>>       ADUNCONST(df1b2vector,x)
>>
>> should have been
>>
>>     ADUNCONST(dvector,x)
>>
>>
>>
>> df1b2vector operator + (const dvector& _x,const df1b2vector& _y)
>>   {
>>     ADUNCONST(df1b2vector,x)
>>
>>    ADUNCONST(df1b2vector,y)
>>
>>    check_shape(x,y,"operator +");
>>
>>
>>    int mmin=x.indexmin();
>>
>>    int mmax=x.indexmax();
>>
>>    df1b2vector tmp;
>>
>>    tmp.noallocate(mmin,mmax);
>>
>>
>>    for (int i=mmin;i<=mmax;i++)
>>     {
>>
>>      tmp(i)=x(i)+y(i);
>>     }
>>
>>    return tmp;
>>
>>  }
>>
>>
>> Sorry for that.  The reason for these stupid errors is that every once
>> and a while the  compiler police for the various c++ compilers make
>> changes that break huge amount of old code so one has to fix it all at
>> one go and these sorts of errors  creep in.
>>
>> Maybe someone can comment on how this can get fixed quickly.
>>
>> as for just hoping that prob will remain between 0 and 1 -- that is a
>> mugs game. It will break at the most inopportune time. I modifed your
>> code to put in the bounds. also you need to include the prior for the
>> u's
>>
>> Finally WRT yuo commentabout initial values for the u's being necessary.
>> that is inaccurate. default values are provided for all parameters.
>> However if you want to supply a *.pin ifle then it must include values
>> for all parameters. I have already commented on the easiest way to
>> produce such a *.pin file.
>>
>> Herre is your code as I have modified it.
>>
>>  //cout << "cvec calc\n";
>>  cvec = sigma_c*(Z*u)+c;
>>  a1vec = elem_prod(pow(size/d,g),exp(-size/d));
>>  //cout << "prob calc\n";
>>  prob = 1/(1/(elem_prod(cvec,a1vec))+h*initial);
>>
>>  dvariable ppen=0.0;
>>  dvar_vector pprob=posfun(prob,0.01,ppen);
>>  f+=ppen;
>>
>>  ppen=0.0;
>>
>>  dvar_vector pprob2=1.0-posfun(1.0-pprob,0.01,ppen);
>>
>>  f+=ppen;
>>
>>  f+=0.5*norm2(u);  // prior
>>
>>  f -= sum( log_comb(initial,killed)+
>>           elem_prod(killed,log(pprob2))+
>>           elem_prod(initial-killed,log(1-pprob2)));
>>
>>
>>
>>
>>
>>   
> 


-- 
Ben Bolker
Associate professor, Biology Dep't, Univ. of Florida
bolker at ufl.edu / www.zoology.ufl.edu/bolker
GPG key: www.zoology.ufl.edu/bolker/benbolker-publickey.asc

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
URL: <http://lists.admb-project.org/pipermail/users/attachments/20090820/1d9a8a13/attachment.pgp>