pmat
a parameter that can be optimized?What is a parallel accumulator?
Points to make
TRUE
:has_age(params)
has_testing(state = state)
has_vax(params)
has_zeta(params)
-- although I am not actually sure if this will always be offdo_step
is now entirely in C++, so I do not understand opt_pars
on page 10 of the vignettepfun
be enough for now? Check to see where/how pfun
is used in MacPan currently -- No, it will not be enough as we also need to compute the functions of the parameterspmat
for age structure?)Hi Steve,
When I am running MacPan for PHAC, the parameters I am estimating are just beta0 and its relative values (i.e. piecewise time-varying relative scales at different time points where a policy was in place or release). Right now I am only calibrating to daily reported cases, so estimating betas are sufficient for now.
I only have vaccination in the model and NO age structure or testing status. Irena developed the stratification, but she doesn't really run them.
I don't know how much effort it is to optimize Irena's stratification code, because it was not written in an optimized way. It was written in a convenient way for her and for her to debug.
I would recommend sitting down and draw out the design/architect of MacPan before heading in different rabbit holes to repatch with better code. Also making this more modular would help. Let me know how I can help and always happy to sit in these meetings.
Mike
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.