soil profile (Aida's work) and evaluation metrics

[Bijan55699]

Addition of two new soil profiles (Aida's work) and activation of evaluation period command.

[Bijan55699]

soil profile (Aida's work) and evaluation metrics

[huard]

Unless there is a problem with the original formulation, this should not be changed.

This is the definition of the EvaluationPeriod command:

@dataclass
class EvaluationPeriod(RavenCommand):
   """:EvaluationPeriod [period_name] [start yyyy-mm-dd] [end yyyy-mm-dd]"""

   name: str
   start: dt.date
   end: dt.date

   template = ":EvaluationPeriod {name} {start} {end}"

   def to_rv(self):
       return self.template.format(**asdict(self))

In practice, use something like (untested):

model.config.rvt.evaluation_periods = [EvaluationPeriod('wet', dt,date(1990, 1, 1), dt.date(2000, 1, 1))]

[Bijan55699]

@huard @Zeitsperre : Just to be more specific: the original formulation doesn't write the "Evaluation Period" command in the .rvi file, and therefore, the specified evaluation period is not considered in the simulation. The ":EvaluationPeriod {evaluation_periods}" seems to activate this command.
This PR is about the evaluation period and not metrics (sorry for my mistake!)

Thanks

Mohammad

[huard]

What do you mean by "original formulation" ?

RavenPy/ravenpy/config/commands.py

Line 80 in 427f7ca

template = ":EvaluationPeriod {name} {start} {end}"

If there's a but with the EvaluationPeriod configuration, let's fix it in another PR.

[huard]

@Zeitsperre Seems to be a problem with the netcdf build. No objection pulling 3.7 from the suite if that simplifies things.

[Zeitsperre]

I believe it's this issue here: Unidata/netcdf4-python#1179. The wheel is bad on PyPI. I can add the fix for it in here at some point.

[Zeitsperre]

@huard Python3.7 is rapidly approaching end of life, so let's do that and see if the issue stems from there. I'm going to drop it in another PR.