search

CDCgov / cfa-epinow2-pipeline

https://cdcgov.github.io/cfa-epinow2-pipeline/
Apache License 2.0
10 stars 2 forks source link

Document PMF generation choices #29

Open zsusswein opened 1 month ago

zsusswein commented 1 month ago

Not in scope for this PR, but I think we need to do a better job documenting and explaining our choice to always discretize the distribution ourselves and pass the pmf

_Originally posted by @kgostic in https://github.com/CDCgov/cfa-epinow2-pipeline/pull/26#discussion_r1752921040_

athowes commented 1 month ago

As in alternative could be to take parameters and discretise within modelling functions here? It would be less information to store to do it that way. But would also need to know the distribution. And also would be less flexible as to other ways to get PMFs (that aren't a particular workflow with a distribution).

zsusswein commented 1 month ago

Adam, in light of the work in cfa-parameter-estimates what are your thoughts on what should go here vs. there? Have they changed at all?

My preference would be to push documentation upstream to cfa-parameter-estimates where possible.

athowes commented 1 month ago

cfa-parameter-estimates is not public and the path to making it public is not easy. Not a complete answer but just to say if you want documentation to be public it's going to be a while before it can fully be pushed upstream. But yes eventually that's my preference too.

zsusswein commented 1 month ago

@kgostic -- thoughts?