admb-project / admb Goto Github PK

The change will be on August 15, 2016.

http://admb-project-org.admb-foundation.org

am testing creating now on Windows 10 vm

@johnoel I currently cannot build on my Windows x64 machine using Mingw and msys. I was able to build previously. I am not sure what I should do. I have been able to build in the past, and I know that as of 32a2f38 it was working for me. What should I do to try to debug? Any ideas are greatly appreciated. Thanks.

Here is some output from the console of the failed build.
make[5]: Entering directory /c/admb/contrib/ecolib' ../../admb.cmd -c Gompertz.cpp cp Gompertz.obj ../../build/objects/dist/contrib-saflp-Gompertz.obj ../../admb.cmd: line 1: @echo: command not found ../../admb.cmd: line 2: @REM: command not found ../../admb.cmd: line 3: syntax error near unexpected token ('
../../admb.cmd: line 3: @REM Copyright 2013 (c) ADMB Foundation' cp: cannot stat Gompertz.obj': No such file or directory
../admb-rules.mak:27: recipe for target ../../build/objects/dist/contrib-saflp-Gompertz.obj' failed make[5]: *** [../../build/objects/dist/contrib-saflp-Gompertz.obj] Error 1 make[5]: Leaving directory /c/admb/contrib/ecolib'
GNUmakefile:153: recipe for target contrib-ecolib' failed make[4]: *** [contrib-ecolib] Error 2 make[4]: Leaving directory /c/admb/contrib'
GNUmakefile:52: recipe for target all' failed make[3]: *** [all] Error 2 make[3]: Leaving directory /c/admb/contrib'
Makefile:80: recipe for target g++-contrib' failed make[2]: *** [g++-contrib] Error 2 make[2]: Leaving directory /c/admb'
Makefile:71: recipe for target g++-all' failed make[1]: *** [g++-all] Error 2 make[1]: Leaving directory /c/admb'
Makefile:5: recipe for target `all' failed
make: *** [all] Error 2

The mailing list:
http://lists.admb-project.org/pipermail/developers/
needs to get copied somewhere

Should use -force_load instead.

tmp.elem(i)=sqrt(v1.elem(i));

sqrt should be sqr or v1.elem(i) * v1.elem(i)?

11.4 manual doesn't appear to be online.

Also, adding text to describe linkage of "ragged array" jargon with number_vectors and vector_vectors would be useful within the manual (description here is quite light).

The following call to cumd_gamma(x,a) using cumd_gamma(257.812,258.714); results in the following error in user

"a too large, ITMAX too small in routine gser"

Solution seems to be to increase ITMAX = 100 to ITMAX = 200 in dfgammp.cpp and cgamdev.cpp

https://groups.google.com/a/admb-project.org/d/msg/users/pB7SaJXyEp4/s7D3LyYzBgAJ

http://www.openbugs.net/Examples/Volumei.html

I had an issue with building in the home directory "c:~\admb" on a windows 64x machine using mingw and msys. The issue was fixed by moving to "c:\admb". See below for emails between @johnoel and @kellijohnson

On 3/31/2016 11:21 AM, Johnoel Ancheta wrote:
Move or clone the admb directory to
c:\admb
Then
$ cd /c/admb
Try and rebuild
$ make clean
$ make

On Thu, Mar 31, 2016 at 5:42 AM, Kelli Johnson wrote:
I am trying to build admb from source on a windows machine, where I
have cloned the github repo and I have mingw and msys. When I type
make I get the following error:
make[3]: Entering directory '/c/~admb/src'
make --directories=src CC=gcc CXX=g++ SHARED=-shared shared
mkdir -p ../build/objects/dist/
g++ -shared -o ../build/dist/lib/libadmb.so -Wl,--whole-archive
../build/dist/l
ib/libadmb-shared.a -Wl,--no-whole-archive
../build/dist/lib/libadmb-shared.a: member
../build/dist/lib/libadmb-shared.a(sh
ared-saflp-linad99-adglobl.obj) in archive is not an object
collect2.exe: error: ld returned 1 exit status
make[3]: *** [shared] Error 1

For all mfexp functions

the INSTALL link under the section Documentation does not work

https://groups.google.com/a/admb-project.org/d/msg/users/FdQ7OLvqGtw/0xyczZ8uBAAJ

http://stackoverflow.com/questions/3072423/changing-to-remove-path-from-env-variable-path

Hover color not working for the main menu in safari.

Is there any way to calculate the higher order derivatives? Particularly, I want to get the gradient of a function which itself is the gradient of another function.

See https://groups.google.com/a/admb-project.org/forum/#!topic/users/TfPNB1Ehp5E

Also, add unit tests.

http://www.admb-project.org/examples should be in version control.

See https://groups.google.com/a/admb-project.org/forum/#!topic/users/nhtGvezR4Vo

For Cole Monnahan

For Windows, the exports below fail to set the flags properly in the shell..

compile: tiny_ad
tiny_ad: tiny_ad.tpl
export CXXFLAGS=-D_USE_MATH_DEFINES
export LDFLAGS=-lm
admb tiny_ad
export CXXFLAGS=
export LDFLAGS=

For Shareef Siddeek.

The precision of reported objective function value in the .par output file should be increased. See nh99/model.cpp for source code, void initial_params::save()

Does any one depend on the number digits reported in the .par file to generate customized reports?

Work with Cole Monnahan to add in.

http://www.admb-project.org/developers

https://groups.google.com/a/admb-project.org/d/msg/developers/OKSr_YJOKi8/onkpmPBBBAAJ

To reproduce this error, open simple.tpl and add these lines to the DATA_SECTION:

init_int sage
init_int nage
vector age(sage,nage)

This worked fine in previous versions but now gives an ambiguity error in dvector.h. There are two allocate functions which match. They are:

void allocate(int ncl, int ncu);
void allocate(const ad_integer &, const index_type &);

Both those definitions were there in the previous versions and have been since inception so something else has changed in the ADMB source to cause this. The error appeared when I used the current source, pulled and compiled on Nov 4, 2015.

Also redirect www.tmb-project.org to github.com/kaskr/adcomp

Read https://groups.google.com/a/admb-project.org/d/msg/users/XPXH-_F_Rlk/dmECoO3vDgAJ

Reported by Jim Ianelli
Example

File t.tpl

DATA_SECTION
6darray d6(1,1,1,2,1,3,1,4,1,5,1,6);
PARAMETER_SECTION
init_number x;
objective_function_value ff
PROCEDURE_SECTION
ff=x*x;

admb script outputs

$ admb t.tpl

*** Parse: t.tpl
tpl2cpp t || tpl2rem t

Error: could not parse "t.tpl".

warning -- creating unallocated 5darray() at line 2
Error in line 2 while reading
(
Error in line 2 while reading
d
make: *** [all] Error 1

init_vector_vector doesn't accept an integer (scalar) variable for phase with typical error as:

mymod.cpp:52:15: error: no matching member function for call to 'allocate'
  MyRaggedMatrix.allocate(1,n,1,m,iPhase,"MyRaggedMatrix");

d3b4116

http://www.admb-project.org/courses/previous/february-2011 points to the 2013 pdf

http://admb-foundation.org/SanSebastianPhotos/ dead link from http://www.admb-project.org/courses/previous/july-2009

http://www.ices.dk/iceswork/training/AD_modelbuilder_and_stock_assessment.pdf dead link from http://www.admb-project.org/courses/previous/other-courses-no-materials-available/ad-model-builder-and-stock-assessment.-21201325-february-2011-at-ices-hq-in-copenhagen