Skip to content
GitLab
Commit 9f2e2b58 authored by Le Roux Erwan's avatar Le Roux Erwan
Browse files

[TEST][TEMPORAL MARGIN] add non stationary test for margin

parent a1a41ac1
master
No related merge requests found
Hide whitespace changes
Inline Side-by-side
Showing
with 49 additions and 47 deletions
+49 -47
test/test_extreme_estimator/test_extreme_models/test_margin_temporal.py
+ 49
− 47
View file @ 9f2e2b58
...@@ -27,16 +27,14 @@ class TestMarginTemporal(unittest.TestCase): ...@@ -27,16 +27,14 @@ class TestMarginTemporal(unittest.TestCase):
def setUp(self) -> None: def setUp(self) -> None:
set_seed_for_test(seed=42) set_seed_for_test(seed=42)
self.nb_points = 2 self.nb_points = 2
self.nb_steps = 5 self.nb_steps = 50
self.nb_obs = 1 self.nb_obs = 1
# Load some 2D spatial coordinates # Load some 2D spatial coordinates
self.coordinates = load_test_spatiotemporal_coordinates(nb_steps=self.nb_steps, nb_points=self.nb_points)[1] self.coordinates = load_test_spatiotemporal_coordinates(nb_steps=self.nb_steps, nb_points=self.nb_points)[1]
self.start_year = 2 smooth_margin_model = LinearNonStationaryLocationMarginModel(coordinates=self.coordinates,
smooth_margin_models = LinearNonStationaryLocationMarginModel(coordinates=self.coordinates, starting_point=2)
starting_point=self.start_year)
self.dataset = MarginDataset.from_sampling(nb_obs=self.nb_obs, self.dataset = MarginDataset.from_sampling(nb_obs=self.nb_obs,
margin_model=smooth_margin_models, margin_model=smooth_margin_model,
coordinates=self.coordinates) coordinates=self.coordinates)
def test_margin_fit_stationary(self): def test_margin_fit_stationary(self):
...@@ -44,52 +42,56 @@ class TestMarginTemporal(unittest.TestCase): ...@@ -44,52 +42,56 @@ class TestMarginTemporal(unittest.TestCase):
margin_model = LinearStationaryMarginModel(self.coordinates) margin_model = LinearStationaryMarginModel(self.coordinates)
estimator = LinearMarginEstimator(self.dataset, margin_model) estimator = LinearMarginEstimator(self.dataset, margin_model)
estimator.fit() estimator.fit()
ref = {'loc': 2.2985600257321295, 'scale': 8.937484202730161, 'shape': 5.744352285758161} ref = {'loc': 1.3403346591679877, 'scale': 1.054867229157924, 'shape': 0.713700960727747}
for year in range(1, 3): for year in range(1, 3):
coordinate = np.array([0.0, 0.0, year]) coordinate = np.array([0.0, 0.0, year])
mle_params_estimated = estimator.margin_function_fitted.get_gev_params(coordinate).to_dict() mle_params_estimated = estimator.margin_function_fitted.get_gev_params(coordinate).to_dict()
print(mle_params_estimated)
for key in ref.keys(): for key in ref.keys():
self.assertAlmostEqual(ref[key], mle_params_estimated[key], places=3) self.assertAlmostEqual(ref[key], mle_params_estimated[key], places=3)
#
# def test_gev_temporal_margin_fit_nonstationary(self): def test_margin_fit_nonstationary(self):
# # Create estimator # Create estimator
# margin_model = NonStationaryStationModel(self.coordinates) margin_model = LinearNonStationaryLocationMarginModel(self.coordinates)
# estimator = LinearMarginEstimator(self.dataset, margin_model) estimator = LinearMarginEstimator(self.dataset, margin_model)
# estimator.fit() estimator.fit()
# self.assertNotEqual(estimator.margin_function_fitted.mu1_temporal_trend, 0.0) self.assertNotEqual(estimator.margin_function_fitted.mu1_temporal_trend, 0.0)
# # Checks that parameters returned are indeed different # Checks that parameters returned are indeed different
# mle_params_estimated_year1 = estimator.margin_function_fitted.get_gev_params(np.array([1])).to_dict() coordinate1 = np.array([0.0, 0.0, 1])
# mle_params_estimated_year3 = estimator.margin_function_fitted.get_gev_params(np.array([3])).to_dict() mle_params_estimated_year1 = estimator.margin_function_fitted.get_gev_params(coordinate1).to_dict()
# self.assertNotEqual(mle_params_estimated_year1, mle_params_estimated_year3) coordinate3 = np.array([0.0, 0.0, 3])
# mle_params_estimated_year3 = estimator.margin_function_fitted.get_gev_params(coordinate3).to_dict()
# def test_gev_temporal_margin_fit_nonstationary_with_start_point(self): self.assertNotEqual(mle_params_estimated_year1, mle_params_estimated_year3)
# # Create estimator
# estimator = self.fit_non_stationary_estimator(starting_point=3) def test_margin_fit_nonstationary_with_start_point(self):
# self.assertNotEqual(estimator.margin_function_fitted.mu1_temporal_trend, 0.0) # Create estimator
# # Checks starting point parameter are well passed estimator = self.fit_non_stationary_estimator(starting_point=2)
# self.assertEqual(3, estimator.margin_function_fitted.starting_point) print(estimator.margin_function_fitted.mu1_temporal_trend)
# # Checks that parameters returned are indeed different self.assertNotEqual(estimator.margin_function_fitted.mu1_temporal_trend, 0.0)
# mle_params_estimated_year1 = estimator.margin_function_fitted.get_gev_params(np.array([1])).to_dict() # Checks starting point parameter are well passed
# mle_params_estimated_year3 = estimator.margin_function_fitted.get_gev_params(np.array([3])).to_dict() self.assertEqual(2, estimator.margin_function_fitted.starting_point)
# self.assertEqual(mle_params_estimated_year1, mle_params_estimated_year3) # Checks that parameters returned are indeed different
# mle_params_estimated_year5 = estimator.margin_function_fitted.get_gev_params(np.array([5])).to_dict() coordinate1 = np.array([0.0, 0.0, 1])
# self.assertNotEqual(mle_params_estimated_year5, mle_params_estimated_year3) mle_params_estimated_year1 = estimator.margin_function_fitted.get_gev_params(coordinate1).to_dict()
# coordinate2 = np.array([0.0, 0.0, 2])
# def fit_non_stationary_estimator(self, starting_point): mle_params_estimated_year2 = estimator.margin_function_fitted.get_gev_params(coordinate2).to_dict()
# margin_model = NonStationaryStationModel(self.coordinates, starting_point=starting_point + self.start_year) self.assertEqual(mle_params_estimated_year1, mle_params_estimated_year2)
# estimator = LinearMarginEstimator(self.dataset, margin_model) coordinate5 = np.array([0.0, 0.0, 5])
# estimator.fit() mle_params_estimated_year5 = estimator.margin_function_fitted.get_gev_params(coordinate5).to_dict()
# return estimator self.assertNotEqual(mle_params_estimated_year5, mle_params_estimated_year2)
#
# def test_two_different_starting_points(self): def fit_non_stationary_estimator(self, starting_point):
# # Create two different estimators margin_model = LinearNonStationaryLocationMarginModel(self.coordinates, starting_point=starting_point)
# estimator1 = self.fit_non_stationary_estimator(starting_point=3) estimator = LinearMarginEstimator(self.dataset, margin_model)
# estimator2 = self.fit_non_stationary_estimator(starting_point=28) estimator.fit()
# mu1_estimator1 = estimator1.margin_function_fitted.mu1_temporal_trend return estimator
# mu1_estimator2 = estimator2.margin_function_fitted.mu1_temporal_trend
# print(mu1_estimator1, mu1_estimator2) def test_two_different_starting_points(self):
# self.assertNotEqual(mu1_estimator1, mu1_estimator2) # Create two different estimators
estimator1 = self.fit_non_stationary_estimator(starting_point=3)
estimator2 = self.fit_non_stationary_estimator(starting_point=20)
mu1_estimator1 = estimator1.margin_function_fitted.mu1_temporal_trend
mu1_estimator2 = estimator2.margin_function_fitted.mu1_temporal_trend
self.assertNotEqual(mu1_estimator1, mu1_estimator2)
if __name__ == '__main__': if __name__ == '__main__':
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment