gsl25, gsl25-devel

[Predrag Punosevac]

Ifigeneia Apostolopoulou <iapostol at andrew.cmu.edu> wrote:

> thanks alot both of you!
> 
> I replaced some functions to use the older  gsl version
> so that I can finish part of my experiments as a temporary solution
> 

You can put those functions back and run the experiment just like you
did on your MAC.

Josko Plazonic of Princeton University who is the lead developer of
Springdale Linux has built RPMs for both gsl25 and gsl25-devel. It's
already in computational repo and I installed on low1
(gsl25, gsl25-devel, env modules in
/usr/local/share/Modules/modulefiles/...).


I hope you do realize that this software is not well tested and any
crazy things should be reported.

On the related note since I am CC-ing this to the entire lab I am adding
llvm-toolset-7 (clang) to all computing nodes right now and replacing
devtoolset-6 with devtoolset-7 (GCC 7).

Cheers,
Predrag


> 
> 
> On Wed, Oct 10, 2018 at 2:58 PM Predrag Punosevac <predragp at andrew.cmu.edu>
> wrote:
> 
> > I just asked on Springdale mailing lists if anybody has encountered this
> > before and what is the safest way to add GSL 2.5 which was released three
> > months ago.
> >
> > Predrag
> >
> > On Wed, Oct 10, 2018 at 1:44 PM Donghan Wang <donghanw at cs.cmu.edu> wrote:
> >
> >> Ifi,
> >>
> >> The following commands should compile the code on LOW1 with GSL 2.5 or
> >> newer.
> >>
> >> source /opt/rh/devtoolset-7/enable
> >> g++ -g -std=c++11 -Wall Kernels.cpp PointProcess.cpp HawkesProcess.cpp
> >> GeneralizedHawkesProcess.cpp EventSequence.cpp
> >> polyagammaSampler/PolyaGamma.cpp polyagammaSampler/RNG.cpp
> >> polyagammaSampler/GRNG.cpp stat_utils.cpp plot_utils.cpp fit_pp_models.cpp
> >> -lboost_iostreams -lboost_filesystem -lboost_system -lboost_serialization
> >> -lboost_timer -lpthread -lgsl -lcblas -llapack
> >> -I/usr/local/gsl/2.4/x86_64/include/
> >>
> >> Note the path specification needs to contain the trailing slash.
> >>
> >> The commands will fail due to lack of gsl_ran_wishart function. The
> >> function is added to GSL 2.5 however GSL 2.4 is currently installed on
> >> LOW1. Predrag is aware of the issue and will look into it.
> >>
> >> Thanks,
> >> Jarod
> >>
> >> On Wed, Oct 10, 2018 at 1:13 PM Ifigeneia Apostolopoulou <
> >> iapostol at andrew.cmu.edu> wrote:
> >>
> >>>  g++ -g -std=c++11 -Wall Kernels.cpp PointProcess.cpp HawkesProcess.cpp
> >>> GeneralizedHawkesProcess.cpp EventSequence.cpp  polyagammaSampler/PolyaGamma.cpp
> >>> polyagammaSampler/RNG.cpp polyagammaSampler/GRNG.cpp stat_utils.cpp
> >>> plot_utils.cpp gen_synth_data.cpp -lboost_iostreams -lboost_filesystem
> >>> -lboost_system -lboost_serialization  -lboost_timer -lpthread -lgsl
> >>> -lcblas -llapack -o gen
> >>>
> >>>
> >>> ./gen 2 1 dpp_example_1 2  0 500   1.0 5  1.0 5   2 1 0 1 0 1    2 1  0
> >>> 1 0 1
> >>>
> >>> On Wed, Oct 10, 2018 at 1:05 PM Ifigeneia Apostolopoulou <
> >>> iapostol at andrew.cmu.edu> wrote:
> >>>
> >>>>  pplib.zip
> >>>> <https://drive.google.com/a/andrew.cmu.edu/file/d/19Unmt8vztfckrDXrbVik3-ZhNiMOWm4x/view?usp=drive_web>
> >>>>
> >>>> g++ -g -std=c++11 -Wall Kernels.cpp PointProcess.cpp HawkesProcess.cpp
> >>>> GeneralizedHawkesProcess.cpp EventSequence.cpp  polyagammaSampler/PolyaGamma.cpp
> >>>> polyagammaSampler/RNG.cpp polyagammaSampler/GRNG.cpp stat_utils.cpp
> >>>> plot_utils.cpp fit_pp_models.cpp -lboost_iostreams -lboost_filesystem
> >>>> -lboost_system -lboost_serialization  -lboost_timer -lpthread -lgsl
> >>>> -lcblas -llapack
> >>>>
> >>>