Adrian Seyboldt

Oh, and I'm not sure, but doesn't this look biased? 

I don't know for sure, bad step size adaptation might explain what you are seeing. If the mass matrix itself is good, but the final step size doesn't match the...

Thank you for looking into this! It sounds to me like it could be this issue? https://github.com/aesara-devs/aesara/issues/127 This happens in the generated code of pytensor if I understand correctly, maybe...

Does anyone know what `sys.platform` will report in the CI? Because that's what we use to determine if we add the extra compiler flags: https://github.com/aesara-devs/aesara/pull/972/files#diff-abd6d8030e381e908a5a5bb1829e6524f4686e78f00bddfcebc134f55c762b1bL2362

Hi :-) You are right, we could really use some documentation on that. If you can share a notebook that we can add to the docs, that would be great!...

@ricardoV94 Nuts for instance also treats nan as -inf. Without this many more models would diverge. Especially early on during sampling nans are not that uncommon, when we might be...

But then, I guess a big difference is that in NUTS we still record that this happend in the form of a divergence. I don't think there is currently anything...

I guess the underlying reason for the failure is that pickling of `DensityDist` doesn't work out of the box? Sounds like for some reason the dispatch functions don't get registered...

I don't mean that the pickling itself throws an error (it doesn't), but that it would be the responsibility of the `DensityDist` object to ensure that the set-up it needs...

As I noticed in nutpie, it is also surprisingly easy to build a custom progress bar. We could also adapt something like the nutpie one into pymc if that makes...