diff --git a/projects/projected_snowfall/elevation_temporal_model_for_projections/ensemble_fit/clustered_ensemble.py b/projects/projected_snowfall/elevation_temporal_model_for_projections/ensemble_fit/clustered_ensemble.py
new file mode 100644
index 0000000000000000000000000000000000000000..755905142c4fe40cdacc3326aca84b597cf51b68
--- /dev/null
+++ b/projects/projected_snowfall/elevation_temporal_model_for_projections/ensemble_fit/clustered_ensemble.py
@@ -0,0 +1,19 @@
+
+"""Instead of creating one big group, with all the ensemble together,
+and assuming a common temporal trend to all the group.
+
+We could create smaller groups, with an importance proportional to the number of GCM/RCM couples considered
+in the group.
+For instance, we could group them by GCM, or group them by RCM.
+Or we could try to find a metric to group them together.
+
+This is the idea of finding of sweet spot between:
+-only independent fits with few assumptions
+-one common fit with too much assumption
+
+
+it links with the idea of "climate model subset".
+
+Generally people try to find one model subset,
+the idea here, would be to find group of model subsets
+"""
\ No newline at end of file
diff --git a/test/test_projects/test_altitude_spatial/test_one_fold_fit.py b/test/test_projects/test_altitude_spatial/test_one_fold_fit.py
new file mode 100644
index 0000000000000000000000000000000000000000..fa4c9161f066741dd0e668ba43785243cfd12dda
--- /dev/null
+++ b/test/test_projects/test_altitude_spatial/test_one_fold_fit.py
@@ -0,0 +1,52 @@
+import unittest
+
+from extreme_data.meteo_france_data.adamont_data.adamont.adamont_snowfall import AdamontSnowfall
+from extreme_data.meteo_france_data.scm_models_data.safran.safran import SafranSnowfall1Day
+from extreme_fit.model.margin_model.polynomial_margin_model.gev_altitudinal_models import StationaryAltitudinal
+from extreme_fit.model.margin_model.polynomial_margin_model.models_based_on_pariwise_analysis.gev_with_constant_shape_wrt_altitude import \
+    AltitudinalShapeConstantTimeLocationLinear, AltitudinalShapeConstantTimeScaleLinear, \
+    AltitudinalShapeConstantTimeLocScaleLinear
+from projects.altitude_spatial_model.altitudes_fit.altitudes_studies import AltitudesStudies
+from projects.altitude_spatial_model.altitudes_fit.one_fold_analysis.one_fold_fit import OneFoldFit
+from spatio_temporal_dataset.coordinates.temporal_coordinates.abstract_temporal_covariate_for_fit import \
+    TimeTemporalCovariate
+
+
+class TestOneFoldFit(unittest.TestCase):
+    pass
+
+    def setUp(self) -> None:
+        super().setUp()
+        self.altitudes = [1200, 1500, 1800]
+        self.massif_name = "Vanoise"
+        self.model_classes = [StationaryAltitudinal,
+                              AltitudinalShapeConstantTimeLocationLinear,
+                              AltitudinalShapeConstantTimeScaleLinear,
+                              AltitudinalShapeConstantTimeLocScaleLinear
+                              ][:]
+
+    def load_dataset(self, study_class):
+        self.studies = AltitudesStudies(study_class, self.altitudes)
+        dataset = self.studies.spatio_temporal_dataset(massif_name=self.massif_name)
+        return dataset
+
+    def test_without_temporal_covariate(self):
+        for study_class in [SafranSnowfall1Day, AdamontSnowfall][:]:
+            dataset = self.load_dataset(study_class)
+            one_fold_fit = OneFoldFit(self.massif_name, dataset,
+                                      models_classes=self.model_classes, temporal_covariate_for_fit=None)
+            print(type(one_fold_fit.best_estimator.margin_model))
+        self.assertTrue(True)
+
+    def test_with_temporal_covariate_for_time(self):
+        for study_class in [SafranSnowfall1Day, AdamontSnowfall][:]:
+            dataset = self.load_dataset(study_class)
+            one_fold_fit = OneFoldFit(self.massif_name, dataset,
+                                      models_classes=self.model_classes,
+                                      temporal_covariate_for_fit=TimeTemporalCovariate)
+            print(type(one_fold_fit.best_estimator.margin_model))
+        self.assertTrue(True)
+
+
+if __name__ == '__main__':
+    unittest.main()