Open
Description
It might be nice to be able to do the following kinds of things with multivariate distributions:
- Marginalize over a subset of variables.
- Calculate conditional distributions conditioned on a subset.
- Assign names of arbitrary types to these variables instead of non-negative integers. For example, might be nice to be able to do something like
val g = MultivariateGaussian(...)
val mg = g.marginalized(List("A", "D", "E"))
This should give us g
marginalized over the variables named A
, D
, and E
.