PEtab import: fix handling of fixed parameters for rule targets

[dweindl]

Previously, targets of initial assignments or rules could have been selected as fixed parameters, which would lead to errors during model import. Those target parameters should be ignored.

[codecov]

Codecov Report

Merging #1915 (a5d5f85) into develop (45e311b) will decrease coverage by 0.01%.
The diff coverage is 66.66%.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #1915      +/-   ##
===========================================
- Coverage    76.01%   75.99%   -0.02%     
===========================================
  Files           76       76              
  Lines        12957    12960       +3     
===========================================
  Hits          9849     9849              
- Misses        3108     3111       +3     

Flag	Coverage Δ
cpp	73.11% <ø> (-0.03%)	⬇️
petab	59.72% <66.66%> (+<0.01%)	⬆️
python	69.10% <0.00%> (-0.05%)	⬇️
sbmlsuite	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
python/sdist/amici/petab_import.py	65.86% <66.66%> (+<0.01%)	⬆️
src/sundials_matrix_wrapper.cpp	80.61% <0.00%> (-0.41%)	⬇️

[FFroehlich]

Previously, targets of initial assignments or rules could have been selected as fixed parameters, which would lead to errors during model import. Those target parameters should be ignored.

Hmm, but this would be an invalid petab file then, right?

[dweindl]

Hmm, but this would be an invalid petab file then, right?

No, I don't mean that the targets were specified as non-estimated in the parameters table, but that during amici import most non-estimated parameters are turned into fixed parameters for performance reasons. And the current logic tries to also do that for rule-targets, which is wrong. (Problem introduced in #1810)

[FFroehlich]

Hmm, but this would be an invalid petab file then, right?

No, I don't mean that the targets were specified as non-estimated in the parameters table, but that during amici import most non-estimated parameters are turned into fixed parameters for performance reasons. And the current logic tries to also do that for rule-targets, which is wrong. (Problem introduced in #1810)

Ah, I see, so the parameter table doesn't need to account for all parameters in a model and the rest are assumed to remain at their default values, right?

[sonarqubecloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

[dweindl]

Ah, I see, so the parameter table doesn't need to account for all parameters in a model and the rest are assumed to remain at their default values, right?

Correct.