Closed willwerscheid closed 3 years ago
my tests suggest that we should continue to use nlm for point-normal but nohess_nlm should be default for point-Laplace (and point-exponential)
more extensive tests suggest that nohess_nlm is ok for all parametric families
closed by pull request #48
my tests suggest that we should continue to use nlm for point-normal but nohess_nlm should be default for point-Laplace (and point-exponential)