Fix format documentation regarding parameter scale

[dweindl]

No description provided.

[codecov]

Codecov Report

Merging #289 into develop will not change coverage by %.
The diff coverage is n/a.

@@           Coverage Diff            @@
##           develop     #289   +/-   ##
========================================
  Coverage    77.18%   77.18%           
========================================
  Files           22       22           
  Lines         1823     1823           
  Branches       398      398           
========================================
  Hits          1407     1407           
  Misses         320      320           
  Partials        96       96           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 023f19e...41cce5b. Read the comment docs.

[fbergmann]

that works for me, I'm still really, really unhappy, that the new interpretation of parameterScale is incompatible with what we did last year. Last year the meaning was, that if paramterScale was log10, then the corresponding values had to be taken as pow(10, value). You decided to be fine with basically invalidating all implementations from last year.

But that is independent of this commit, this commit is fine and necessary. And it is consistent with the decision you made.

[dweindl]

Yeah, sorry for breaking things there. Apart from the breaking change, I think it makes life easier in the future though. If one only wants to simulate the model, one can simply ignore the parameter scaling, and it put an end to the confusion with the observable transformation which was handled incoherently.