I’ve used this in the past when I was doing more analytics and it worked great. It’s interesting that you bring up the “good documentation” piece now because I credit this package with finally teaching me what is actually going on under the hood with conjugate priors of different distributions versus thinking of Bayesian stats more as “call this function and trust me”.
I’m a big fan of all the documentation and error checking built directly into the package. I seem to remember earlier versions/advertisements billing it only as fast (which it also is ).
Or - or, and just hear me out, there are occasional real world complications that cause unforeseen errors and people only talk about the tail end of poor predictions rather than the 95+% good experiences.
I’m a big fan of all the documentation and error checking built directly into the package. I seem to remember earlier versions/advertisements billing it only as fast (which it also is ).