NaN to nan and numpy<2

sgatto · Sep 24, 2024 · 1efbffa · 1efbffa
1 parent 75e7253
commit 1efbffa
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 5 deletions.
diff --git a/openstef/tasks/create_solar_forecast.py b/openstef/tasks/create_solar_forecast.py
@@ -186,7 +186,7 @@ def fides(data: pd.DataFrame, all_forecasts: bool = False):
     data = pd.DataFrame(index = index,
                         data = dict(load=np.sin(index.hour/24*np.pi)*np.random.uniform(0.7,1.7, 300)))
     data['insolation'] = data.load * np.random.uniform(0.8, 1.2, len(index)) + 0.1
-    data.loc[int(len(index)/3*2):,"load"] = np.NaN
+    data.loc[int(len(index)/3*2):,"load"] = np.nan
 
     """
     insolation_forecast = apply_fit_insol(data, add_to_df=False)
@@ -357,7 +357,7 @@ def apply_fit_insol(data, add_to_df=True, hours_delta=None, polynomial=False):
     data = pd.DataFrame(index = index,
                         data = dict(load=np.sin(index.hour/24*np.pi)*np.random.uniform(0.7,1.7, len(index))))
     data['insolation'] = data.load * np.random.uniform(0.8, 1.2, len(index)) + 0.1
-    data.loc[int(len(index)/3*2):,"load"] = np.NaN
+    data.loc[int(len(index)/3*2):,"load"] = np.nan
 
     """
     colname = list(data)[0]

diff --git a/requirements.txt b/requirements.txt
@@ -7,6 +7,7 @@ lightgbm~=3.3
 matplotlib~=3.7
 mlflow~=2.3
 networkx~=3.1
+numpy<2.0.0
 optuna~=3.1
 optuna-integration~=3.6
 pandas~=2.2.0

diff --git a/test/unit/tasks/test_calculate_kpi.py b/test/unit/tasks/test_calculate_kpi.py
@@ -18,11 +18,11 @@
 
 # Prepare dataframe with nans to test low completeness
 realised_load_nan = realised_load.copy()
-realised_load_nan.loc[realised_load_nan.sample(frac=0.5).index, :] = np.NaN
+realised_load_nan.loc[realised_load_nan.sample(frac=0.5).index, :] = np.nan
 
 # Prepare dataframe with nans to test low completeness
 predicted_load_nan = predicted_load.copy()
-predicted_load_nan.loc[predicted_load_nan.sample(frac=0.5).index, :] = np.NaN
+predicted_load_nan.loc[predicted_load_nan.sample(frac=0.5).index, :] = np.nan
 
 prediction_job = TestData.get_prediction_job(pid=307)
 
@@ -106,7 +106,7 @@ def test_calc_kpi_for_specific_pid_poor_completeness_realized(self):
             prediction_job["id"], realised_load_nan, predicted_load, realised_load_nan
         )
         t_ahead_keys = kpis.keys()
-        self.assertIs(kpis[list(t_ahead_keys)[0]]["rMAE"], np.NaN)
+        self.assertIs(kpis[list(t_ahead_keys)[0]]["rMAE"], np.nan)
 
     # Test whether none is returned in case of poor completeness for predicted data
     def test_calc_kpi_for_specific_pid_poor_completeness_predicted(self):