Issue 561: Soft min transformation #562
Conversation
Also removed unnecessary call to `fetch`
Try this Pull Request!Open Julia and type: import Pkg
Pkg.activate(temp=true)
Pkg.add(url="https://github.com/CDCgov/Rt-without-renewal", rev="new-obs-trans", subdir="EpiAware")
using EpiAware |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #562 +/- ##
=======================================
Coverage 90.96% 90.96%
=======================================
Files 60 60
Lines 863 863
=======================================
Hits 785 785
Misses 78 78 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Nice this looks good to me! Phew we should have done this a while ago.
Note I rebased this to your target PR vs main to catch this is a stacked PR.
If fetch was from Dagger does this mean Dagger is living in the Project.toml still and not being used? Should we remove it?
Dagger isn't in the Project anymore, I missed this in testing because |
* Create make_mcmc_diagnostic_dataframe.jl * reorg scripts and add more success/fail analysis * Add function to get run info to avoid DRY * Add function to do diagnostics * export new func * update SI * Issue 561: Soft min transformation (#562) Also removed unnecessary call to `fetch` * base values on pipeline types * breakdown mcmc convergence test function Adds more stats and a unit test
This PR closes #561.
This is a small PR which simply redefines the
Ascertainmentpart of the observation model composition to have a custom transformation which soft mins the expected observations at1e15chosen so that theInexactErrorsare not caused by wild sampling in some models.I also increased the pipeline test draws to better check if this is operating as expected.
PS
I've done this as an extension of the work in #560 so if @seabbs is happy with that we can do this PR as a 2-fer.
Minor change: I noted that
fetchwas still being used in the tests as a legacy of usingDagger.jl. This didn't error because it has a sensibleBasedefault but I've removed as unnecessary.