Bizzi
Bizzi
Indeed, nested differentiation of closures has been a [well known problem for quite some time](https://juliadiff.org/ReverseDiff.jl/limits/) . In multiple places it is mentioned that a fix is on the works. Are...
Hey. Not only do Distributions still behave differently with regards to types when generating one or multiple samples ( #1252 ), but the behavior is also inconsistent between different types:...
Hey. Great package. Trying to implement some coupling layers, I noticed that some architectures combining `Parallel` and `WrappedFunctions` can be type stable when the parameters are `NamedTuple`s but not when...