Automatic imputation - array dimension problem

[ckrapu]

I recommend reviewing the discussion at these threads:

Multivariate normal with missing data imputation operands could not be broadcast together with shapes (29,2) () (29,)

Try: pip install -U git+https://github.com/arviz-devs/arviz.git pip install -U git+https://github.com/pymc-devs/pymc3.git

How to marginalize on a Multivariate Normal distribution

I am new to both probabilistic modeling and PyMC so please bear with me. I have a model that learns a Multivariate normal distribution and now I’d like to find the probabilities over a partial test input. My problem is very similar to the one described in this question. Specifically, I refer to the final comment quoted as below: you need to work out the conditional probability of ma. by marginalizing the MvNormal using [.5, 1.2] and cov, and generate from this new conditional probability (al…

In particular, the first thread recommends using trace = pm.sample(return_inferencedata=False, compute_convergence_checks=False) so that the postprocessing doesn’t execute.