More generic posterior sampling
It was mentioned that the machinery for drawing posterior samples should be made more general, and should not only be available for
I'd be happy to implement this, but I need a clear description of the required ingredients and the algorithm.
Concerning this functionality, there are currently several open questions:
- issue #211 (closed)
- issue #200 (closed)
- NIFTy4 currently has two implementations:
generate_posterior_sample2(), both in
nifty4/library/wiener_filter_curvature.py. Which is the correct one?