[Feature request]: demonstrate a model selection process using flepimop

[pearsonca]

Label

meta/workflow

Priority Label

medium priority

Is your feature request related to a problem? Please describe.

Users routinely do trial-and-error, ad hoc model selection. However, formal methods for model selection could be more effective.

Is your feature request related to a new application, scenario round, pathogen? Please describe.

No response

Describe the solution you'd like

Currently, doing model selection looks like

$ flepimop calibrate someconfigA.yml # creates inference results
$ flepimop simulate someconfigA,yml # creates time series, or maybe also outputs of inference
$ jupyter render notebook.ipynb # and then look at results
$ flepimop simulate someconfigB.yml # and then tweak notebook
$ jupyter render notebook.ipynb # and then look at results
$ # repeat as necessary

a formal model selection pipeline would look like

$ flepimop calibrate someconfig.yml # some config contains multiple models to fit, A, B, C, etc
$ flepimop process someconfig.yml --target=select # has a process module/definition 'select' that does model selection
$ flepimop process someconfig.yml --target=display # has a process module/definition 'display' that renders model selection analysis outcomes