Add possibility to CircuitTemplate.apply() method to change constants

[pyrates-neuroscience]

Add possibility to CircuitTemplate.apply() method to change constants

[dafrose]

Funny... I kind of expected to work already, but it doesn't. All other templates already support that functionality. Thus, it is only a matter having the keyword/dictionary point to the respective node/edge and so on. For nodes it should be easy. For edges it is necessary to specify source and target as in the template and it might still not be unique, if multiple edges between the same nodes are defined in a circuit template. I do not think, there is an easy way around this (except for overwriting the edges in a custom yaml/Python template).

Thus the partial solution: Implement node_values keyword argument, that can point to nodes by key as defined in the template. The rest is simply passed to the node template's apply method. I would prefer to leave edges the way they are now, because this could potentially introduce bugs leading to unexpected results (or confuse error handling). Would that be sufficient for now?

[Richert]

I think that sounds like a good temporary solution. If we have a good idea for how to handle changes to edge attributes during the CircuitTemplate.apply() call, we can easily add this later on.

[dafrose]

Implemented with 1d4428e . Behaviour is documented in test_frontend_yaml_parser.py/test_circuit_instantiation. I went for a dictionary of the form {'node/op/var':value} for consistency with existing APIs. Closing.