put 'method' in class definition Pt.2

Author: thierrymoudiki

GPopt/GPOpt.py

@@ -858,7 +858,6 @@ def lazyoptimize(
         abs_tol=None,  # suggested 1e-4, for n_iter = 200
         min_budget=50,  # minimum budget for early stopping
         func_args=None,
-        method="bayesian",  # "bayesian" or "mc
         estimators="all",
         type_pi="kde", # for now, 'kde', 'bootstrap', 'splitconformal'
         type_exec="independent",  # "queue" or "independent" (default)
@@ -884,9 +883,6 @@ def lazyoptimize(
             func_args: a list;
                 additional parameters for the objective function (if necessary)
 
-            method: an str;
-                "bayesian" (default) for Gaussian posteriors or "mc" for Monte Carlo posteriors
-
             estimators: an str or a list of strs (estimators names)
                 if "all", then 30 models are fitted. Otherwise, only those provided in the list
                 are adjusted; for example ["RandomForestRegressor", "Ridge"]
@@ -978,6 +974,7 @@ def lazyoptimize(
                 seed=self.seed,
                 n_jobs=self.n_jobs,
                 acquisition=self.acquisition,
+                method=self.method,
                 min_value=self.min_value,
                 surrogate_obj=copy.deepcopy(self.regressors[0][1]),
             )
@@ -987,7 +984,6 @@ def lazyoptimize(
                 abs_tol=abs_tol,  # suggested 1e-4, for n_iter = 200
                 min_budget=min_budget,  # minimum budget for early stopping
                 func_args=func_args,
-                method=method,
             )
 
             score_next_param = gp_opt_obj_prev.y_min
@@ -1014,6 +1010,7 @@ def lazyoptimize(
                             seed=self.seed,
                             n_jobs=self.n_jobs,
                             acquisition=self.acquisition,
+                            method=self.method,
                             min_value=self.min_value,
                             surrogate_obj=copy.deepcopy(self.regressors[i][1]),
                             x_init=np.asarray(gp_opt_obj_prev.parameters),
@@ -1099,6 +1096,7 @@ def lazyoptimize(
                             seed=self.seed,
                             n_jobs=self.n_jobs,
                             acquisition=self.acquisition,
+                            method=self.method,
                             min_value=self.min_value,
                             surrogate_obj=copy.deepcopy(self.regressors[i][1]),
                         )
@@ -1148,6 +1146,7 @@ def foo(i):
                         seed=self.seed,
                         n_jobs=self.n_jobs,
                         acquisition=self.acquisition,
+                        method=self.method,
                         min_value=self.min_value,
                         surrogate_obj=copy.deepcopy(self.regressors[i][1]),
                     )

examples/lazybranin.py

@@ -1,8 +1,10 @@
 import os 
 import GPopt as gp
+import nnetsauce as ns 
 import numpy as np
 from os import chdir
 from scipy.optimize import minimize
+from sklearn.linear_model import Ridge 
 
 print(f"\n ----- Running: {os.path.basename(__file__)}... ----- \n")
 
@@ -54,4 +56,7 @@ def branin(x):
                    n_iter=190,                    
                    seed=4327)
 
+print(f"gp_opt1.method: {gp_opt1.method}")                   
+
 gp_opt1.lazyoptimize(verbose=2, abs_tol=1e-4)
+