fixed get best results for agg=True, gpyopt allow reps flag fixed, gpyopt repeat easy test

Author: larshertel

sherpa/algorithms/bayesian_optimization.py

@@ -49,7 +49,7 @@ class GPyOpt(Algorithm):
         verbosity (bool): Print models and other options during the optimization.
         max_num_trials (int): maximum number of trials to run for.
     """
-    allows_repetition = False
+    allows_repetition = True
 
     def __init__(self, model_type='GP', num_initial_data_points='infer',
                  initial_data_points=[], acquisition_type='EI',
@@ -131,7 +131,6 @@ def _num_completed_trials(cls, results):
 
     def _generate_bayesopt_batch(self, X, y, lower_is_better, domain):
         y_adjusted = y * (-1)**(not lower_is_better)
-
         bo_step = gpyopt_package.methods.BayesianOptimization(f=None,
                                                               domain=domain,
                                                               X=X, Y=y_adjusted,

sherpa/algorithms/core.py

@@ -154,7 +154,8 @@ def get_best_result(self, parameters, results, lower_is_better):
                                              parameters=parameters,
                                              min_count=self.num_times)
 
-        print(agg_results)
+        if agg_results.empty:
+            return {}
 
         # Get best result so far
         best_idx = (agg_results.loc[:, 'Objective'].idxmin()

tests/test_algorithms.py

@@ -473,6 +473,24 @@ def test_repeat_get_best_result():
     assert study.get_best_result()['a'] == 1  # not 3
 
 
+def test_repeat_get_best_result_called_midway():
+    parameters = [sherpa.Choice('a', [1,2,3])]
+    gs = sherpa.algorithms.GridSearch()
+    gs = sherpa.algorithms.Repeat(algorithm=gs, num_times=3)
+    study = sherpa.Study(parameters=parameters, algorithm=gs,
+                         lower_is_better=True,
+                         disable_dashboard=True)
+
+    objectives = [2.1,2.2,2.3, 9., 0.1, 9.1, 1.1,1.2,1.3]
+    expected = [None, None, 1, 1, 1, 1, 1, 1, 3]
+
+    for exp, obj, trial in zip(expected, objectives, study):
+        study.add_observation(trial, objective=obj)
+        study.finalize(trial)
+        assert study.get_best_result().get('a') == exp
+
+
+
 def test_repeat_results_aggregation():
     parameters = [sherpa.Continuous('myparam', [0, 1])]
 

tests/test_gpyopt.py

@@ -368,41 +368,39 @@ def obj_func(x, y, z):
     assert rval['z'] == 5
 
 
-# def test_noisy_parabola():
-#     def f(x, sd=1):
-#         y = (x - 3) ** 2 + 10.
-#         if sd == 0:
-#             return y
-#         else:
-#             return y + numpy.random.normal(loc=0., scale=sd,
-#                                            size=numpy.array(x).shape)
-#
-#     parameters = [sherpa.Continuous('x1', [0., 7.])]
-#
-#     bayesian_optimization = GPyOpt(max_concurrent=1,
-#                                    max_num_trials=20,
-#                                    model_type='GP',
-#                                    acquisition_type='EI')
-#     rep = Repeat(algorithm=bayesian_optimization,
-#                  num_times=5)
-#     study = sherpa.Study(algorithm=rep,
-#                          parameters=parameters,
-#                          lower_is_better=True,
-#                          disable_dashboard=True)
-#
-#     for trial in study:
-#         print("Trial {}:\t{}".format(trial.id, trial.parameters))
-#
-#         fval = f(trial.parameters['x1'], sd=1)
-#         print("Function Value: {}".format(fval))
-#         study.add_observation(trial=trial,
-#                               iteration=1,
-#                               objective=fval)
-#         study.finalize(trial, status='COMPLETED')
-#     rval = study.get_best_result()
-#     print(rval)
-#     # assert numpy.sqrt((rval['Objective'] - 3.)**2) < 0.2
-
-#
-# if __name__ == '__main__':
-#     test_noisy_parabola()
+def test_noisy_parabola():
+    def f(x, sd=1):
+        y = (x - 3) ** 2 + 10.
+        if sd == 0:
+            return y
+        else:
+            return y + numpy.random.normal(loc=0., scale=sd,
+                                           size=numpy.array(x).shape)
+
+    parameters = [sherpa.Continuous('x1', [0., 7.])]
+
+    bayesian_optimization = GPyOpt(max_concurrent=1,
+                                   max_num_trials=5,
+                                   model_type='GP',
+                                   acquisition_type='EI')
+    rep = Repeat(algorithm=bayesian_optimization,
+                 num_times=3,
+                 agg=True)
+    study = sherpa.Study(algorithm=rep,
+                         parameters=parameters,
+                         lower_is_better=True,
+                         disable_dashboard=True)
+
+    for trial in study:
+        # print("Trial {}:\t{}".format(trial.id, trial.parameters))
+
+        fval = f(trial.parameters['x1'], sd=1)
+        # print("Function Value: {}".format(fval))
+        study.add_observation(trial=trial,
+                              iteration=1,
+                              objective=fval)
+        study.finalize(trial, status='COMPLETED')
+    # rval = study.get_best_result()
+    # print(rval)
+    print(study.results.query("Status=='COMPLETED'"))
+    # assert numpy.sqrt((rval['Objective'] - 3.)**2) < 0.2