Semi-open uniform distribution


Newbie here, hello. Is there a way to create a semi-open uniform distribution. One can have pm.Uniform(a, b), but I’d like to use something like pm.Uniform(a, np.inf) or pm.Uniform(-np.inf, b).

I suppose I could mangle in a way of using a pm.HalfFlat, but it seems like a pm.HalfOpenFlat would be useful, and more general, than HalfFlat.


It will be an improper distribution anyway, so I would just do a + pm.HalfFlat(...).