Refactoring (part 1) #18
Conversation
|
launch jenkins |
1 similar comment
|
launch jenkins |
|
launch jenkins |
1 similar comment
|
launch jenkins |
|
All tests are green. Ready to review 💡 |
| # Nudging (meteorological and tracers) | ||
| era5_global_nudging = False | ||
| species_global_nudging = False | ||
| species2nudge = [] |
There was a problem hiding this comment.
Are there future plans of doing something with this?
There was a problem hiding this comment.
@jthanwer What do you think?
There was a problem hiding this comment.
Does species2nudge have any impact on the results or is this is only a feature to be implemented in the future?
| 'oem_gridded_emissions_nc': ['tno_3cat.nc', 'OEM'], | ||
| 'oem_vertical_profiles_nc': ['vertical_profiles.nc', 'OEM'], | ||
| 'oem_hourofday_nc': ['hourofday.nc', 'OEM'], | ||
| 'oem_dayofweek_nc': ['dayofweek.nc', 'OEM'], | ||
| 'oem_monthofyear_nc': ['monthofyear.nc', 'OEM'], |
There was a problem hiding this comment.
Am I correct in seeing that this specification for OEM differs between COSMO and ICON?
There was a problem hiding this comment.
True, but it could be adapted for COSMO to have it more uniform.
There was a problem hiding this comment.
Yeah, I have to say, it's actually quite neat. I'd support for COSMO following a similar quickfire settings procedure. If @gredvis agrees.
There was a problem hiding this comment.
Yes, I agree that it would be nice to have the same look-and-feel for COSMO
|
@efmkoene Thanks for your review. Indeed, the ensembles cases is not included here, but it will come back as a test case in the next iteration (see description of this PR). Let me know if you are happy with the changes. |
|
launch jenkins |
|
I briefly had a look at it together with Erik. Looks all good to me. I propose to move forward with the merge. |
|
Sorry, I should not have closed the pull request |
Updating and consolidating Processing Chain for ICON-ART (C2SM Task WG2 in 2023/1)
With the increasing number of supported model configurations and simulation options, parts of the processing chain (especially the main routine
run_chain.py) have become rather complex and somewhat convoluted.The following parts of refactoring the code are addressed in this PR:
target->model,subtarget->variantconfig.yamlfile)models.yamlfile for global model configurationsrun_chain.py: distinguish features and variants instead of modelsThis is part 1 of the refactoring task. Part 2 will mainly consist of the following:
config.py->config.yamlvariantsandfeaturescan be unified