Let’s discuss what steps would be required to extend the current GP api in some meaningful ways, given the upcoming
MatrixNormal distribution. Some possibilities include
Coregionkernels (as in GPflow or GPy, which uses a mixed noise likelihood. Does that have any benefit over
- Multiple curve observations of same GP (works without
MatrixNormal, but needs testing)
- I personally enjoy
plot_gp_distand would like to see if it could be modified to handle the above points
- Other ideas?
These features would help in my research, so this is partly a selfish plea for your help, but I really do think that PyMC3 would benefit from these additions. This post is just meant to spur discussion on these points.