In the (preprint) paper on page 8, I find that
Each posterior uses 4096 samples, or 256 for ABC; all MAP estimates are found using the mean-shift algorithm.
Can I find hints to this MAP estimate in this repo? If not so, which implementation did you use for the plots?
In the (preprint) paper on page 8, I find that
Can I find hints to this MAP estimate in this repo? If not so, which implementation did you use for the plots?