ensemble estimators test (#50)

test/common_test.py

@@ -113,4 +113,3 @@ def test_classifier_cv(self):
                         yp = eq.predict_proba(self.X_test)
                         np.testing.assert_equal(y.shape, self.y_test.shape)
                         np.testing.assert_equal(yp.shape, (self.y_test.shape[0], 2))
-

test/ensemble_test.py

@@ -0,0 +1,82 @@
+from HROCH import SymbolicRegressor, NonlinearLogisticRegressor, SymbolicClassifier, FuzzyRegressor, FuzzyClassifier
+import unittest
+from sklearn.datasets import load_breast_cancer
+from sklearn.model_selection import train_test_split
+from sklearn.ensemble import BaggingRegressor, BaggingClassifier, VotingRegressor, VotingClassifier, StackingRegressor, StackingClassifier, AdaBoostRegressor, AdaBoostClassifier
+
+
+class TestEnsemble(unittest.TestCase):
+    def __init__(self, *args, **kwargs):
+        super(TestEnsemble, self).__init__(*args, **kwargs)
+
+        self.params = {'num_threads': 1, 'time_limit': 0.0,'iter_limit': 1000, 'random_state': 42}
+        X, y = load_breast_cancer(return_X_y=True)
+        self.X_train, self.X_test, self.y_train, self.y_test = train_test_split(
+            X, y, test_size=0.2, random_state=self.params['random_state'])
+
+    def test_bagging_regressor(self):
+        ensemble_model = BaggingRegressor(estimator=SymbolicRegressor(**self.params))
+        ensemble_model.fit(self.X_train, self.y_train)
+        y_pred = ensemble_model.predict(self.X_test)
+        self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_adaboost_regressor(self):
+        ensemble_model = AdaBoostRegressor(estimator=SymbolicRegressor(**self.params))
+        ensemble_model.fit(self.X_train, self.y_train)
+        y_pred = ensemble_model.predict(self.X_test)
+        self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_stacking_regressor(self):
+        base_model = SymbolicRegressor(**self.params)
+        base_model.fit(self.X_train, self.y_train)
+        math_models = base_model.get_models()[:5]
+        estimators = [(str(m), m) for m in math_models]
+        ensemble_model = StackingRegressor(estimators=estimators)
+        ensemble_model.fit(self.X_train, self.y_train)
+        y_pred = ensemble_model.predict(self.X_test)
+        self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_voting_regressor(self):
+        base_model = SymbolicRegressor(**self.params)
+        base_model.fit(self.X_train, self.y_train)
+        math_models = base_model.get_models()[:5]
+        estimators = [(str(m), m) for m in math_models]
+        ensemble_model = VotingRegressor(estimators=estimators)
+        ensemble_model.fit(self.X_train, self.y_train)
+        y_pred = ensemble_model.predict(self.X_test)
+        self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_bagging_classifier(self):
+        for c in [NonlinearLogisticRegressor, FuzzyRegressor]:
+            ensemble_model = BaggingClassifier(estimator=NonlinearLogisticRegressor(**self.params))
+            ensemble_model.fit(self.X_train, self.y_train)
+            y_pred = ensemble_model.predict(self.X_test)
+            self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_adaboost_classifier(self):
+        ensemble_model = AdaBoostClassifier(estimator=NonlinearLogisticRegressor(**self.params))
+        ensemble_model.fit(self.X_train, self.y_train)
+        y_pred = ensemble_model.predict(self.X_test)
+        self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_stacking_classifier(self):
+        for c in [NonlinearLogisticRegressor, FuzzyRegressor]:
+            base_model = c(**self.params)
+            base_model.fit(self.X_train, self.y_train)
+            math_models = base_model.get_models()[:5]
+            estimators = [(str(m), m) for m in math_models]
+            ensemble_model = StackingClassifier(estimators=estimators)
+            ensemble_model.fit(self.X_train, self.y_train)
+            y_pred = ensemble_model.predict(self.X_test)
+            self.assertEqual(y_pred.shape[0], self.y_test.shape[0])
+
+    def test_voting_classifier(self):
+        for c in [NonlinearLogisticRegressor, FuzzyRegressor]:
+            base_model = c(**self.params)
+            base_model.fit(self.X_train, self.y_train)
+            math_models = base_model.get_models()[:5]
+            estimators = [(str(m), m) for m in math_models]
+            ensemble_model = VotingClassifier(estimators=estimators)
+            ensemble_model.fit(self.X_train, self.y_train)
+            y_pred = ensemble_model.predict(self.X_test)
+            self.assertEqual(y_pred.shape[0], self.y_test.shape[0])