Seth Axen
Seth Axen
Bump, maybe @devmotion or @torfjelde?
Bump again.
Additionally, and this is more of an AbstractMCMC comment, if `sample` could take an array of models and not just a model, then we could also use the `MCMCThreads`, `MCMCDistributed`,...
Yes, actually, diagnosing issues with SBI for posterior inference is one of the use cases I have for SBC.
Yeah, SBC is also quite different than the diagnostics here in that a PPL-agnostic implementation does almost nothing, while a PPL-specific implementation would be really convenient but would essentially require...
> I don't think this has to or should be couple to a PPL. For instance, for algorithm 2 it seems one just needs to provide functions that sample from...
How would one pass a vector of draws to e.g. `ess_rhat`?
There are two different ways to think about this. One is to reason about what permutation users are likely to pass (which leads to reasoning about what a sensible ordering...
I agree it makes sense to keep base behavior here. It seems like a macro that overrides base behavior and uses dimensions would be useful. I guess a generalization of...
> Thinking about this we should probably start by adding a warning for these things. Agreed with this! RE options 1-4, I'm still thinking through whether these would cause any...