search

stephenslab / ebnm

R package to fit Empirical Bayes Normal Means model.
https://stephenslab.github.io/ebnm/
12 stars 9 forks source link

add tests for very large and small data #40

Closed willwerscheid closed 3 years ago

willwerscheid commented 3 years ago

we haven't sufficiently tested scaling

willwerscheid commented 3 years ago

It doesn't do well for large x due to the stepmax setting for nlm (which I need to avoid overflow errors). I think I'll have to add pre- and postscaling functions so that optimization is done over (for example) x / median(s)

willwerscheid commented 3 years ago

after scaling, set default stepmax (nlm) to 1 and rmax (trust) to 1; otherwise they fail too easily