TST: complementing tests for sensitivity analysis and removing duplicate piece of code.

Author: Lucas-Prates

rocketpy/sensitivity/sensitivity_model.py

@@ -140,10 +140,6 @@ def set_target_variables_nominal(self, target_variables_nominal_value):
             self.target_variables_info[target_variable]["nominal_value"] = (
                 target_variables_nominal_value[i]
             )
-        for i, target_variable in enumerate(self.target_variables_names):
-            self.target_variables_info[target_variable]["nominal_value"] = (
-                target_variables_nominal_value[i]
-            )
 
         self._nominal_target_passed = True
 
@@ -356,12 +352,12 @@ def __check_requirements(self):
             version = ">=0" if not version else version
             try:
                 check_requirement_version(module_name, version)
-            except (ValueError, ImportError) as e:
+            except (ValueError, ImportError) as e:  # pragma: no cover
                 has_error = True
                 print(
                     f"The following error occurred while importing {module_name}: {e}"
                 )
-        if has_error:
+        if has_error:  # pragma: no cover
             print(
                 "Given the above errors, some methods may not work. Please run "
                 + "'pip install rocketpy[sensitivity]' to install extra requirements."

tests/unit/test_sensitivity.py

@@ -1,13 +1,11 @@
+from unittest.mock import patch
+
 import numpy as np
 import pytest
 
 from rocketpy.sensitivity import SensitivityModel
 
-# TODO: for some weird reason, these tests are not passing in the CI, but
-#       passing locally. Need to investigate why.
-
 
-@pytest.mark.skip(reason="legacy test")
 def test_initialization():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1", "target2"]
@@ -21,7 +19,6 @@ def test_initialization():
     assert not model._fitted
 
 
-@pytest.mark.skip(reason="legacy test")
 def test_set_parameters_nominal():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1", "target2"]
@@ -35,8 +32,16 @@ def test_set_parameters_nominal():
     assert model.parameters_info["param1"]["nominal_mean"] == 1.0
     assert model.parameters_info["param2"]["nominal_sd"] == 0.2
 
+    # check dimensions mismatch error raise
+    incorrect_nominal_mean = np.array([1.0])
+    with pytest.raises(ValueError):
+        model.set_parameters_nominal(incorrect_nominal_mean, parameters_nominal_sd)
+
+    incorrect_nominal_sd = np.array([0.1])
+    with pytest.raises(ValueError):
+        model.set_parameters_nominal(parameters_nominal_mean, incorrect_nominal_sd)
+
 
-@pytest.mark.skip(reason="legacy test")
 def test_set_target_variables_nominal():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1", "target2"]
@@ -49,9 +54,13 @@ def test_set_target_variables_nominal():
     assert model.target_variables_info["target1"]["nominal_value"] == 10.0
     assert model.target_variables_info["target2"]["nominal_value"] == 20.0
 
+    # check dimensions mismatch error raise
+    incorrect_nominal_value = np.array([10.0])
+    with pytest.raises(ValueError):
+        model.set_target_variables_nominal(incorrect_nominal_value)
+
 
-@pytest.mark.skip(reason="legacy test")
-def test_fit_method():
+def test_fit_method_one_target():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1"]
     model = SensitivityModel(parameters_names, target_variables_names)
@@ -65,7 +74,20 @@ def test_fit_method():
     assert model.number_of_samples == 3
 
 
-@pytest.mark.skip(reason="legacy test")
+def test_fit_method_multiple_target():
+    parameters_names = ["param1", "param2"]
+    target_variables_names = ["target1", "target2"]
+    model = SensitivityModel(parameters_names, target_variables_names)
+
+    parameters_matrix = np.array([[1.0, 2.0], [2.0, 3.0], [3.0, 4.0]])
+    target_data = np.array([[10.0, 12.0, 14.0], [11.0, 13.0, 17.0]]).T
+
+    model.fit(parameters_matrix, target_data)
+
+    assert model._fitted
+    assert model.number_of_samples == 3
+
+
 def test_fit_raises_error_on_mismatched_dimensions():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1"]
@@ -78,7 +100,6 @@ def test_fit_raises_error_on_mismatched_dimensions():
         model.fit(parameters_matrix, target_data)
 
 
-@pytest.mark.skip(reason="legacy test")
 def test_check_conformity():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1", "target2"]
@@ -90,7 +111,6 @@ def test_check_conformity():
     model._SensitivityModel__check_conformity(parameters_matrix, target_data)
 
 
-@pytest.mark.skip(reason="legacy test")
 def test_check_conformity_raises_error():
     parameters_names = ["param1", "param2"]
     target_variables_names = ["target1", "target2"]
@@ -101,3 +121,42 @@ def test_check_conformity_raises_error():
 
     with pytest.raises(ValueError):
         model._SensitivityModel__check_conformity(parameters_matrix, target_data)
+
+    parameters_matrix2 = np.array([[1.0, 2.0, 3.0], [2.0, 3.0, 4.0]])
+
+    with pytest.raises(ValueError):
+        model._SensitivityModel__check_conformity(parameters_matrix2, target_data)
+
+    target_data2 = np.array([10.0, 12.0])
+
+    with pytest.raises(ValueError):
+        model._SensitivityModel__check_conformity(parameters_matrix, target_data2)
+
+    target_variables_names = ["target1"]
+    model = SensitivityModel(parameters_names, target_variables_names)
+
+    target_data = np.array([[10.0, 20.0], [12.0, 22.0], [14.0, 24.0]])
+
+    with pytest.raises(ValueError):
+        model._SensitivityModel__check_conformity(parameters_matrix, target_data)
+
+
+@patch("matplotlib.pyplot.show")
+def test_prints_and_plots(mock_show):  # pylint: disable=unused-argument
+    parameters_names = ["param1", "param2"]
+    target_variables_names = ["target1"]
+    model = SensitivityModel(parameters_names, target_variables_names)
+
+    parameters_matrix = np.array([[1.0, 2.0], [2.0, 3.0], [3.0, 4.0]])
+    target_data = np.array([10.0, 12.0, 14.0])
+
+    # tests if an error is raised if summary is called before print
+    with pytest.raises(ValueError):
+        model.info()
+
+    model.fit(parameters_matrix, target_data)
+    assert model.all_info() is None
+
+    nominal_target = np.array([12.0])
+    model.set_target_variables_nominal(nominal_target)
+    assert model.all_info() is None