Batchrunfixedparameters

[Corvince]

Added ability to include fixed parameters in get_model_vars_dataframe() function. See #422

[coveralls]

Coverage decreased (-0.2%) to 81.929% when pulling a0b0158 on Corvince:BatchrunFixedParameters into 275674d on projectmesa:master.

[coveralls]

Coverage decreased (-0.2%) to 81.929% when pulling a0b0158 on Corvince:BatchrunFixedParameters into 275674d on projectmesa:master.

[TaylorMutch]

@Corvince is there any inherit issue with the fixed_parameters being included by default, and have folks opt out of automatically including the fixed parameters? And if the someone doesn't include fixed_parameters as part of the BatchRunner's constructor, then there's no difference.

Also, since _prepare_report_table is a 'private' method anyway, could we instead have the BatchRunner class instead have a method that knows whether to include the fixed parameters or not?

[TaylorMutch]

@Corvince would the following approach achieve the same effect?

In BatchRunner.__init__:

...
self.fixed_parameters = fixed_parameters or {}
self._include_fixed = len(self.fixed_parameters.keys()) > 0
...

In BatchRunner._prepare_report_table:

...
if self._include_fixed:
        for param in self.fixed_parameters.keys():
        ordered[param] = self.fixed_parameters[param]
return ordered
...

[coveralls]

Coverage increased (+0.09%) to 82.911% when pulling 59edb4e on Corvince:BatchrunFixedParameters into 63f3fbd on projectmesa:master.

[Corvince]

Thanks @TaylorMutch. That seems to me like a better approach and achieves the same result. I think it makes sense to have fixed parameters in by default.

[TaylorMutch]

Looks good to me!