You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
On the first iteration of the optimization when using the GP_UCB acquisition function the iteration is 0. This causes nt to be 0 which causes std::log(nt * pi2 / delta3) to be NaN. This results in all values returned by operator() also being NaN for this iteration. Looking at this reference seems to imply that the iteration (referred to as t in the paper) should start at 1 rather than 0.
The text was updated successfully, but these errors were encountered:
On the first iteration of the optimization when using the GP_UCB acquisition function the
iteration
is 0. This causesnt
to be 0 which causesstd::log(nt * pi2 / delta3)
to beNaN
. This results in all values returned byoperator()
also being NaN for this iteration. Looking at this reference seems to imply that the iteration (referred to ast
in the paper) should start at 1 rather than 0.The text was updated successfully, but these errors were encountered: