pysamoo icon indicating copy to clipboard operation
pysamoo copied to clipboard

Default number of DOE points for SurrogateAssistedAlgorithm class

Open AngusKenny opened this issue 1 year ago • 0 comments

In pysamoo.core.algorithm, the docstring for the SurrogateAssistedAlgorithm class states that if n_initial_doe = None, "the default is 11n - 1 (but at most n_max_doe)".

Further down, in _setup(), it sets the number of initial DOE points to: self.n_initial_doe = min(self.n_initial_max_doe, default_n_doe(problem.n_var))

Then at the top of pysamoo.core.algorithm it has the definition for default_n_doe as:

def default_n_doe(n, max=float("inf")):
    return min(2 * n + 1, max)

This suggests the number of initial DOE points would default to 2n + 1, instead of 11n - 1. Is the docstring incorrect or is default_n_doe defined incorrectly? I am using this code in my research, and I just want to be sure that I am using the correct defaults.

Thanks!

AngusKenny avatar Jun 07 '24 02:06 AngusKenny