Issue #443 clipping msw model #1478
Conversation
| # grid is independent of model grid and we do not want to perform | ||
| # transformations on meteo data in this case. | ||
| if has_meteogrid_copy and isinstance(pkg, MeteoMapping): | ||
| clipped[key] = deepcopy(pkg) |
There was a problem hiding this comment.
Is this necessary? You have a clip_box method in the MeteoMapping class. I would expect either that method to be called here or that that method isn't necessary
There was a problem hiding this comment.
The special-case is only necessary when a MeteoGridCopy is assigned to the model. This object just copies a pointer file mete_grid.inp which contains paths to a lot of meteo ASCII files, which are commonly very large. In this case, we do not want to clip these large datasets, nor create mappings to clipped meteo grids (as we do not want to clip these). This is a very common usecase for iMOD5 models.
The MeteoMapping.clip_box method should be called when a MeteoGrid object is assigned to the model.
| return indexer, repeat_stress | ||
|
|
||
|
|
||
| def clip_time_slice( |
There was a problem hiding this comment.
I like that you refactored out duplicate code and reduced it to this single method
| timestamps with data. | ||
| """ | ||
| # First, "pop" and filter. | ||
| keys, values = repeat_stress.values.T |
There was a problem hiding this comment.
You could rename keys to times or period
imod/common/utilities/clip.py
Outdated
| keep = (keys >= time_start) & (keys <= time_end) | ||
| new_keys = keys[keep] | ||
| new_values = values[keep] | ||
| # Now detect which "value" entries have gone missing |
There was a problem hiding this comment.
The code belonging to this code doesn't look for missing entries but for unique entries right?
The comment confuses me
There was a problem hiding this comment.
I've clarified this function with better naming and comments, as it was very confusing
imod/common/utilities/clip.py
Outdated
|
|
||
| # Update the key-value pairs. Discard keys that have been "promoted". | ||
| keep = np.isin(new_keys, insert_keys, assume_unique=True, invert=True) | ||
| new_keys = new_keys[keep] |
There was a problem hiding this comment.
Isn't new_keys equal to insert_keys? and new_values to insert_values?
There was a problem hiding this comment.
I've clarified this function with better naming and comments, as it was very confusing
|
Fixes #443 and #364
Description
imod.msw.MetaSwapModel, this includes a special case forclip_boxmethod of PrecipitationMapping, and EvapotranspirationMapping: These should not be clipped if aMeteoGridCopyinstance is part of the model. Otherwise nonsensical mappings are generated.imod.msw.MetaSwapModel.clip_boxmethod to public APIclip_boxmethod in MeteoMapping, should clip meteo attribute instead of dataset attribute.imod.common.utilities.clip.meteoclass variable an attribute. This makes more sense. I think I made it a class variable before to please mypy.imod.msw.MeteoGridCopyChecklist
Issue #nr, e.g.Issue #737