Feature species - more adjoint parts#1461
Conversation
Generalization of the adjont output necessary
Actually I just noticed there is no |
|
@TobiKattmann I am actually developing the adjoint stuff for nemo right now! Though there are some recording issues.... |
Ok @WallyMaier if this PR gets merged then there will be a |
TobiKattmann
left a comment
There was a problem hiding this comment.
Really nice 👍 puts this logic into one place and avoids a bunch of work for future changes
|
I will add surface avg species_0 as another Objective function (Edit: done in 31d7a95) but that would be it from my side concerning this PR |
Only Species_0 because one can of course always shift the species functions around. In case more than one avg surf Spec is needed at once, the one has to add it manually
|
So the merging process would of course be:
... in case the people are happy with it |
Co-authored-by: Pedro Gomes <pcarruscag@gmail.com>
Note that the AddHistoryOutput lines are read by the update_historyMap.py script. And if there are line breaks there are information missing for that script. Using // clang-format off/on can prevent this accidentally breaking.
Co-authored-by: Pedro Gomes <38071223+pcarruscag@users.noreply.github.com>
Hi,
Some necessary changes for the species transport solver to work nicely with the Discrete adjoint solver. So it is the same structure that was Introduced for CFlowOutput but now it is CAdjFlowOutput.
Related Work
directly following #1388 where the majority of adjoint work for the species solver was already done
PR Checklist