Conversation
…e of the settings) and fix lint
…have different interfaces; also use to enable passing different physics options to SIQN timestepper. Do not expect Tom to like...
tommbendall
added
enhancement
| no_normal_flow_bc_ids (list, optional): a list of IDs of domain | ||
| boundaries at which no normal flow will be enforced. Defaults to | ||
| None. | ||
| """ |
There was a problem hiding this comment.
I think the kwargs are a good solution. We should probably also add something to docstring here about what happens to the kwargs and give some examples of the kwargs that can be passed
| if family is None: | ||
| extruded_mesh = hasattr(mesh, "_base_mesh") | ||
| if extruded_mesh: | ||
| cell = mesh._base_mesh.ufl_cell().cellname() |
There was a problem hiding this comment.
Note that this will be cellname for future
| elif cell == "triangle": | ||
| family = "BDM" | ||
| else: | ||
| raise ValueError(f"The mesh provided (or its base mesh if 3D) must have cells of type interval, quadrilateral or triangle, not {cell}.") |
There was a problem hiding this comment.
Petty point but could this be over 2 lines?
| ) | ||
|
|
||
|
|
||
| class Model(SIQNModel): |
There was a problem hiding this comment.
I had wondered if this should be DefaultModel? Maybe that's not the best name, but Model feels more like the name of a base class? (I was actually a bit confused by the role this played but you haven't had chance to do the docstrings yet!)
| _diffusion_schemes = [] | ||
| for field, _ in self.diffusion_options: | ||
| _diffusion_schemes.append( | ||
| BackwardEuler(self.domain, field) |
There was a problem hiding this comment.
I think this will also need the recovery options specifying
2 participants
No description provided.