[Refactor]: Call REPORT_F from model.hpp instead of fleet and population

[Andrea-Havron-NOAA]

Refactor request

Currently, derived values are reported from fleet and population. When there is more than one fleet or population, however, the vector of derived values will be overwritten. REPORT_F should be called from model.hpp instead to avoid these issues.

Expected behavior

A better structure will be to declare matrices within model.hpp, fill in the matrices within population and fleet loops and report these out directly from model.hpp. There are current issues with calling REPORT_F from model.hpp that need to be resolved.

Code of Conduct

• I agree to follow this project's Code of Conduct

[Andrea-Havron-NOAA]

I added a call to REPORT_F in model.hpp on the 377-branch that added reporting to model.hpp to recreate the error message.

See the compilation error on the GHA

[msupernaw]

@Andrea Havron - NOAA Federal ***@***.***> There error is because model.hpp uses type T for the template parameter. The code in interface.hpp <https://github.com/NOAA-FIMS/FIMS/blob/377-refactor-call-report_f-from-modelhpp-instead-of-fleet-and-population/inst/include/interface/interface.hpp#L28C3-L28C3> uses type Type in the isDouble<Type> call, which is a preprocessing macro. We'll either need to change the template parameter in model.hpp to Type or isDouble<Type> to isDouble<T> for it to compile.

…

On Tue, Jul 25, 2023 at 6:49 PM Andrea-Havron-NOAA ***@***.***> wrote: I added a call to REPORT_F in model.hpp <https://github.com/NOAA-FIMS/FIMS/blob/c869e979010088fcd53ca320552a7ab3b4783003/inst/include/common/model.hpp#L111> on the 377-branch <https://github.com/NOAA-FIMS/FIMS/tree/377-refactor-call-report_f-from-modelhpp-instead-of-fleet-and-population> that added reporting to model.hpp to recreate the error message. See the compilation error <https://github.com/NOAA-FIMS/FIMS/actions/runs/5661981351/job/15341006819#step:6:62> on the GHA — Reply to this email directly, view it on GitHub <#377 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFUSECUE2O5SLFY2CUT2MDXSBEQVANCNFSM6AAAAAAZNRWLTY> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Matthew Supernaw *Scientific Software Developer* *National Oceanic and Atmospheric Administration* *Office Of Science and Technology* *NOAA Fisheries | *U.S. Department of Commerce Phone 248 - 396 - 7797

[ChristineStawitz-NOAA]

I'm adding a P1 label to this because we report the threshold for accuracy in the FIMS manuscript and it would be good to have a standard value (needing standard errors) before we submit that.

[Andrea-Havron-NOAA]

@timjmiller, I've used your recommendation of vector< vector > to call ragged arrays for reporting out different fleet and population derived values.

To do:

• Fix tests
• Do we want to reassembled dimension folded vectors back into matrices (ie. year x age) for reporting?