[ ] Implement SALib sensitivity analysis in pymob. This should be very easy. The script damage-proxy/case_studies/reversible_damage/scripts/sensitivity_analysis.py is basically self contained. As a module connected to a Simulation object it should already do everything necessary.
[ ] Exploit model runtime information. The progressbar shows that specific parameter combinations have better or worse runtime. This can be exploited to show regions where the ODE model is not well behaved.
[ ] I can use these information perfectly for creating parameter scan visualization (e.g. 1-D or 2-D marginal plots for the influence of parameters on the model solution or the evaluation runtime). Maybe even interactively.
Basically this is a grid search.
Multichain computation on remote kernels with post-chain combination
Parameter analysis
[ ] this is also basically fully implemented in the damage-proxy/case_studies/reversible_damage/scripts/parameter_identifiability.py
SALib sensitivity analysis
SALibsensitivity analysis in pymob. This should be very easy. The scriptdamage-proxy/case_studies/reversible_damage/scripts/sensitivity_analysis.pyis basically self contained. As a module connected to aSimulationobject it should already do everything necessary.Multichain computation on remote kernels with post-chain combination
Parameter analysis