search

nmfs-ost / ss3-test-models

Stock Synthesis models used for testing
Creative Commons Zero v1.0 Universal
3 stars 1 forks source link

update models to clean up some warnings #52

Closed e-perl-NOAA closed 1 year ago

e-perl-NOAA commented 1 year ago

Solves some warnings documented in issue #51.

Note: vermillion_snapper_F4 model still has a warning that suggests starting with a higher R0. I changed the R0 init several times to 11, 13, 15, and 20 (I changed the max too for 15 and 20) but the warning still persisted. @iantaylor-NOAA if you have any suggestions, please let me know.

Rick-Methot-NOAA commented 1 year ago

I will look at the parameter values for selectivity in the vermillion example.

Rick-Methot-NOAA commented 1 year ago

I ran vermillion_F4 locally and do not get that warning when I use 3.30.22. That was because starter.ss said start from ss.par. When I change to start from control file, then I get the warning. The warning gets triggered because the FMethod 4 was starting F from parameters beginning in phase 1, where they have uninformed starting values, so produce big catch logL, triggering the warning. Better set up for FMethod 4 is to delay parameters to later stage for the catch fleets:

(A) (B) (C) (terminate list with -9999 for fleet) 1 0.01 2 # CM_E 2 0.01 2 # CM_W 3 0.01 2 # REC 4 0.01 1 # SMP_BYC -9999 1 1 # end of list