To know how good a PPL is, try writing down: 1, a Categorical regression, check with result when replace Categorical with Multinomial(1, p). 2, a multivariate mixture model, check its forward random generation and how easy to get evaluation of logp for each mixture component.
-
-
I have a lot of examples in PyMC3 showing how we failed

-
It would be a valuable service to the rest of us if you found time to write those up! (Not that I expect you to. Just that I'd be appreciative.)
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.