You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We need to implement a way to conceal the chains too!
This will require us to move the smokescreen namespace to have something like: import smokescreen.ConcealDataVector and import smokescreen.ConcealPosterior to conceal at different levels.
The text was updated successfully, but these errors were encountered:
What do we do with parameters that are prior bounded? these can give away any shifts (@jessmuir has an implementation from the DES analysis for these cases)
Multi-chain support
Two different shifts:
absolute shifts $\Delta \theta$ centered at zero
relative shifts with relation to a fiducial value or a chain such that model comparison can be done (@jessmuir has some examples of how to do this).
Include also an "exclusion list" to avoid shifting some desired parameters
We need to implement a way to conceal the chains too!
This will require us to move the smokescreen namespace to have something like:
import smokescreen.ConcealDataVector
andimport smokescreen.ConcealPosterior
to conceal at different levels.The text was updated successfully, but these errors were encountered: