Issue #1158 remaining hfb bottlenecks #1159
Conversation
JoerivanEngelen
changed the base branch from
master
to
imod5_converter_feature_branch
| @@ -1378,8 +1378,5 @@ def from_imod5_data( | |||
| ) | |||
| simulation["ims"] = solution | |||
|
|
|||
| # cleanup packages for validation | |||
| idomain = groundwaterFlowModel.domain | |||
| simulation.mask_all_models(idomain) | |||
There was a problem hiding this comment.
there are a few examples on regridding. Even if they don't need this masking, it may be helpful to modify them to do it anyway to make the reader aware how to mask all the regridded packages
There was a problem hiding this comment.
The hondsrug example is the example where we regrid a simulation. In this example data is already made consistent, and regridding doesn't introduce any problems. I defer from adding unnecessary calls to the example, as it might lead users to blindly call computationally intensive functions. However, I want to create an example how to import a (problematic) iMOD5 model, where some cleanup needs to be done (for which cleanup utilities have to be made). We could introduce that here. I'll create a separate issue for that.
There was a problem hiding this comment.
See issue here #1164
|
JoerivanEngelen
merged commit ccf9f21
into
imod5_converter_feature_branch
Fixes #1158
Description
Fix remaining HFB bottlenecks. This reduces writing the HFB package from the LHM from 12.5 minutes to 2 minutes.
xu.Ugrid2d.from_structured, as this is a costly operationandinstead of&operator in thescalar_Nonefunction, to enable shortcutting.mask_all_packagesfunction infrom_imod5_data, call in tests.pixi run unittestsstill runs all unittests, by starting two pixi tasksunittests_njit&unittests_jit.Checklist
Issue #nr, e.g.Issue #737