Bayesian Inference with Physical Models


Struggling with the setup of the models.

Here is the current problem I am trying to build:
There is set of 100 prior inputs that are uniform distributed.
These prior inputs are fed into a physical model that spits out an estimated value.
This estimated value is then compared against known test data which has a set mu and sigma.

I want to know what the posterior distribution of the inputs is.

How would this be built in PyMC?

Any help is appreciated.

You can have a look at this discussion: Updating posteriors for scanning a chemical system