tensiometer.synthetic_probability
This module contains all the methods to build synthetic models for posterior distributions.
These models start from samples from a given posterior distribution and build machine learning normalizing flow models for the distribution.
The synthetic distribution can then be evaluated at arbitrary points, is differentiable and we can sample from it.
Note that documentation is spotty at places and might need to be improved.
Core modules