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
I have been trying to use this package with nonlinear-optimization and ad for maximum likelihood computations, however this is highly non-practical since distributions always return Double -> Double and ad needs some kind of polymorphic interface to work. Are there any barriers for such a polymorphic interface?
One problem I see is the need for different contraints depending on the distribution (Erf for Normal, Num for others).
The text was updated successfully, but these errors were encountered:
Unfortunately there are. Mostly because some distributions use functions like (incomplete) beta/gamma functions. Current implementations in math-functions are specialized to Doubles and making them generic is a lot of work.
To make matter more complicated there are also efficiency concerns. For example for incomplete beta/gamma functions derivative exist in closed form and could be calculated very cheaply without using all complicated approximations.
I have been trying to use this package with
nonlinear-optimization
andad
for maximum likelihood computations, however this is highly non-practical since distributions always returnDouble -> Double
andad
needs some kind of polymorphic interface to work. Are there any barriers for such a polymorphic interface?One problem I see is the need for different contraints depending on the distribution (
Erf
for Normal,Num
for others).The text was updated successfully, but these errors were encountered: