modAL.models.BayesianOptimizer.get_max() method added for obtaining the currently best values

Author: cosmic-cortex

modAL/models.py

@@ -459,6 +459,23 @@ def _set_max(self, y):
         if y_max > self.max_val:
             self.max_val = y_max
 
+    def get_max(self):
+        """
+        Gives the highest value so far.
+
+        Returns
+        -------
+        X: np.ndarray of shape (n_features)
+            The location of the currently best value.
+
+        y: np.ndarray of shape (1)
+            The currently best value.
+
+        """
+        max_idx = np.argmax(self.y_training)
+
+        return self.X_training[max_idx], self.y_training[max_idx]
+
     def teach(self, X, y, bootstrap=False, only_new=False, **fit_kwargs):
         """
         Adds X and y to the known training data and retrains the predictor with the

tests/core_tests.py

@@ -588,6 +588,19 @@ def test_set_new_max(self):
                 learner._set_max(y_new)
                 np.testing.assert_almost_equal(np.max(y_new), learner.max_val)
 
+    def test_get_max(self):
+        for n_samples in range(1, 100):
+            for max_idx in range(0, n_samples):
+                X = np.random.rand(n_samples, 3)
+                y = np.random.rand(n_samples)
+                y[max_idx] = 10
+
+                regressor = mock.MockEstimator()
+                optimizer = modAL.models.BayesianOptimizer(regressor, X_training=X, y_training=y)
+                X_max, y_max = optimizer.get_max()
+                np.testing.assert_equal(X_max, X[max_idx])
+                np.testing.assert_equal(y_max, y[max_idx])
+
     def test_teach(self):
         for bootstrap, only_new in product([True, False], [True, False]):
             # case 1. optimizer is uninitialized