Help with mixture model of MvNormals in pymc3?

[junpenglao]

@leka0024

My input space in this toy example only has 3 locations (in one spatial dimension), so it’s not really high dimension. Unless you’re talking the size of the model (# of Thetas)? I don’t understand why the sampler would not have the trace for those “extra” Thetas (without measurements) be 60% of one component and 40% of the other, instead of what looks like 99.9% of one component and 0.1% of the other.

for Mixture distribution, >1 is already a high dimension as the space becomes too difficult to explore for HMC.

@leka0024

This must be the core issue, but I don’t understand what this means. Can a different sampler do this in pymc3, other than NUTS? Or can you suggest if I can try a different package altogether (Stan?), although I really prefer to stay with pymc3 since I’m comfortable in it!

You can try SMC. Any package that use NUTS (Stan included) will give you the same issue.