I am trying to create a mixture model with hyper parameters. Eventually, mu for the two parts will be computed using different equations but for the moment they are the same, while I resolve my problem with label switching. My model is defined as: with pm.Model() as model: coef = pm.Normal('coe…

Constraining order in a Mixture model

AustinRochford August 5, 2021, 11:37am 4

You want to add transform=pm.transforms.ordered to the constructor for coef. See the examples in Chapter 11 of the PyMC3 port of Statistical Rethinking.

1 Like

Topic		Replies	Views
Mixture models - pm.Potential or pm.transforms.ordered version agnostic modeling	2	583	March 25, 2022
Constrain one variable to be greater than another Questions	14	3519	January 30, 2018
Label switching in multivariate mixtures Questions	0	765	September 28, 2020
Why does `transform=pm.distributions.transforms.ordered` lead to worse convergence? Questions	3	1579	August 27, 2021
Mixture models and breaking class symmetry Questions	11	3932	October 12, 2018

Constraining order in a Mixture model

Related topics