changed pd version, fixed benchmark model

pyforecaster/forecasting_models/benchmarks.py

@@ -79,7 +79,7 @@ def __init__(self, period='1d', q_vect=None, val_ratio=None, nodes_at_step=None,
     def fit(self, x:pd.DataFrame, y:pd.DataFrame):
         self.n_sa = y.shape[1]
         # infer sampling time
-        sampling_time = pd.infer_freq(x.index)
+        sampling_time = x.index.drop_duplicates().diff().median()
 
         # retrieve support of target distribution
         support = np.unique(y.values)
@@ -109,11 +109,11 @@ def fit(self, x:pd.DataFrame, y:pd.DataFrame):
         return self
 
     def predict(self, x, **kwargs):
-        return (self.predict_probabilities(x) * np.tile(self.support.reshape(1, -1), self.n_sa)).groupby(level=0, axis=1).sum()
+        return (self.predict_probabilities(x) * np.tile(self.support.reshape(1, -1), self.n_sa)).T.groupby(level=0).sum()
 
     def predict_probabilities(self, x, **kwargs):
         # infer sampling time
-        sampling_time = pd.infer_freq(x.index)
+        sampling_time = x.index.drop_duplicates().diff().median()
         # Create a new column for the time within a day (hours and minutes)
         time_of_day = pd.Series(x.index.floor(sampling_time).time)
 

requirements.txt

@@ -1,7 +1,7 @@
 numpy>=1.20.2
 optuna>=2.10.0
 networkx>=2.6.3
-pandas>=1.2.3
+pandas>=2.2.3
 seaborn>=0.11.1
 matplotlib>=3.4.1
 scipy>=1.7.1

setup.py

@@ -22,7 +22,7 @@
     install_requires=['numpy>=1.20.2',
                       'optuna>=2.10.0',
                       'networkx>=2.6.3',
-                      'pandas>=1.2.3',
+                      'pandas>=2.2.3',
                       'seaborn>=0.11.1',
                       'matplotlib>=3.4.1',
                       'scipy>=1.7.1',