The suggested range for the number of estimators is from 1 to 60. Why did you use np.unique(np.logspace(0, 1.8, num=30)).astype(int)
to derive the range instead of a simple np.arange? Also, this expression produces some repetitions so we have for example [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, …etc.].
The np.unique
will remove the repetitions.
I assume the idea was to sample with a log uniform instead of a uniform distribution to have more chance to get small number of estimators.