jacobnzw / SSMToybox

Nonlinear Sigma-Point Kalman Filters based on Bayesian Quadrature
MIT License
12 stars 0 forks source link

Using `RandomVariable`s in `StateSpaceInference` ? #33

Open jacobnzw opened 3 years ago

jacobnzw commented 3 years ago

The statistics during inference can get pretty confusing when writing the inference machinery, especially in StudentianInference, where one can often loose track of what DoF parameter belongs to which scale matrix.

Explore the option of using subclasses of RandomVariable to group together the statistics during inference. Write an example code to see what it would look like.