fixes #382

Author: brendan

bayes_opt/bayesian_optimization.py

@@ -246,12 +246,8 @@ def _prime_subscriptions(self):
     def maximize(self,
                  init_points=5,
                  n_iter=25,
-                 acq='ucb',
-                 kappa=2.576,
-                 kappa_decay=1,
-                 kappa_decay_delay=0,
-                 xi=0.0,
-                 **gp_params):
+                 utility_function=None):
+
         """
         Probes the target space to find the parameters that yield the maximum
         value for the given function.
@@ -291,13 +287,16 @@ def maximize(self,
         self._prime_subscriptions()
         self.dispatch(Events.OPTIMIZATION_START)
         self._prime_queue(init_points)
-        self.set_gp_params(**gp_params)
 
-        util = UtilityFunction(kind=acq,
-                               kappa=kappa,
-                               xi=xi,
-                               kappa_decay=kappa_decay,
-                               kappa_decay_delay=kappa_decay_delay)
+        if utility_function is None:
+            util = UtilityFunction(kind='ucb',
+                                   kappa=2.576,
+                                   xi=0.0,
+                                   kappa_decay=1,
+                                   kappa_decay_delay=0)
+        else:
+            util = utility_function
+
         iteration = 0
         while not self._queue.empty or iteration < n_iter:
             try:

examples/test.py

@@ -27,10 +27,7 @@ def black_box_function(x, y):
 
         optimizer.maximize(init_points=2,
                      n_iter=25,
-                     acq='ucb',
-                     kappa=5,
-                     xi=1,
-                     **{'normalize_y':True, 'alpha':2.5e-3, 'n_restarts_optimizer':20})
+                     utility_function=utility)
         # for point in range(20):
         #     next_point_to_probe = optimizer.suggest(utility)
         #     NextPointValues = np.array(list(next_point_to_probe.values()))