kernel = "matern", nu = 5/2 is not working

[pranasblk]

`> OPT_Res <- BayesianOptimization(Test_Fun,

• bounds = list(x = c(1, 3)),
• init_points = 2, n_iter = 10,
• acq = "ei", kappa = 2.576, eps = 0.0,
• kernel = "matern", nu = 5/2, verbose = TRUE)`

Result: Error in corr$type : $ operator is invalid for atomic vectors In addition: Warning message: In GPfit::GP_fit(X = Par_Mat[Rounds_Unique, ], Y = Value_Vec[Rounds_Unique], : nug_thres is outside of the normal range of [10, 25].

or

`> OPT_Res <- BayesianOptimization(Test_Fun,

• bounds = list(x = c(1, 3)),
• init_points = 2, n_iter = 10,
• acq = "ei", kappa = 2.576, eps = 0.0,
• kernel = "matern", verbose = TRUE)`

Result: elapsed = 0.00 Round = 1 x = 1.2185 Value = 1.0471 elapsed = 0.00 Round = 2 x = 2.2607 Value = 1.3450 Error in corr$type : $ operator is invalid for atomic vectors

[yanyachen]

kernel parameter should be a list that specifies the type of correlation function along with the smoothness parameter. Popular choices are square exponential (default) or matern 5/2. For example, kernel = list(type = "matern", nu = 5/2)

[pranasblk]

Thanks for looking into it - my mistake. I've just digged it from the GPfit doc.