Generalise modification of parameters in NeuroMLController

[sanjayankur31]

In:

https://github.com/NeuroML/pyNeuroML/blob/development/pyneuroml/tune/NeuroMLController.py#L260

we currently use a process where we hard code certain attributes that can be modified here during the tuning process, but it should be possible to just use an xpath like form that allows modification of any attributes.

That way we don't need to hard-code anything here

[Ragzz1995]

Hi. since I'm new to contributing to opensource projects, I thought I'll start off something with "good first issue tags". So, I hope it's okay if ask some rudimentary questions. By parameters do you mean the parameters passed to this function?

     def __init__(
        self,
        ref,
        neuroml_file,
        target,
        sim_time=1000,
        dt=0.05,
        simulator="jNeuroML",
        generate_dir="./",
        num_parallel_evaluations=1,
        cleanup=True,
    ):

Do you want to parse parameters from xml file using Xpath instead of hardcoding?

[sanjayankur31]

Hi @Ragzz1995

Thanks for the comment.

No, at the moment, the path string used for to pass parameters to the run_optimisation function has a specific format, based on what we've hard-coded in the NeuroMLTuner, and we were wondering if we need to hard-code these or if we can use the usual xpath spec that we use in general with NeuroML. Here's an example code snippet,

https://github.com/NeuroML/Documentation/blob/master/source/Userdocs/NML2_examples/tune-izhikevich.py#L147

and here's where it's "hard-coded":
https://github.com/NeuroML/pyNeuroML/blob/master/pyneuroml/tune/NeuroMLController.py#L318