Fix test_get_metrics_dict_scaler unit-test

tests/gordo/builder/test_builder.py

@@ -9,7 +9,7 @@
 import pandas as pd
 import sklearn.compose
 import sklearn.ensemble
-from sklearn.base import BaseEstimator
+from sklearn.base import BaseEstimator, RegressorMixin
 from sklearn.multioutput import MultiOutputRegressor
 from sklearn.linear_model import LinearRegression
 from sklearn import metrics
@@ -63,6 +63,17 @@ def machine_check(machine: Machine, check_history):
         )
 
 
+class CustomRegressor(BaseEstimator, RegressorMixin):
+    def __init__(self, multiplier):
+        self.multiplier = multiplier
+
+    def fit(self, X, y):
+        return self
+
+    def predict(self, X):
+        return X * self.multiplier
+
+
 @pytest.mark.parametrize("scaler", [None, "sklearn.preprocessing.MinMaxScaler"])
 def test_get_metrics_dict_scaler(scaler):
     metrics_list = [sklearn.metrics.mean_squared_error]
@@ -74,12 +85,10 @@ def test_get_metrics_dict_scaler(scaler):
     metrics_dict = ModelBuilder.build_metrics_dict(metrics_list, y, scaler=scaler)
     metric_func = metrics_dict["mean-squared-error"]
 
-    mock_model1 = MagicMock()
-    mock_model1.predict = lambda _y: _y * [0.8, 1]
-    mse_feature_one_wrong = metric_func(mock_model1, y, y)
-    mock_model2 = MagicMock()
-    mock_model2.predict = lambda _y: _y * [1, 0.8]
-    mse_feature_two_wrong = metric_func(mock_model2, y, y)
+    classifier1 = CustomRegressor(np.array([0.8, 1]))
+    mse_feature_one_wrong = metric_func(classifier1, y, y)
+    classifier2 = CustomRegressor(np.array([1, 0.8]))
+    mse_feature_two_wrong = metric_func(classifier2, y, y)
 
     if scaler:
         assert np.isclose(mse_feature_one_wrong, mse_feature_two_wrong)