epyt-flow 0.6.0__py3-none-any.whl → 0.7.1__py3-none-any.whl

epyt_flow/VERSION

@@ -1 +1 @@
-0.6.0
+0.7.1

epyt_flow/data/benchmarks/leakdb.py

@@ -509,6 +509,7 @@ def load_scenarios(scenarios_id: list[int], use_net1: bool = True,
         if not os.path.exists(f_inp_in):
             with ScenarioSimulator(f_inp_in=network_config.f_inp_in) as wdn:
                 wdn.set_general_parameters(**general_params)
+                wdn.epanet_api.setTimePatternStep(hydraulic_time_step)
 
                 wdn.epanet_api.deletePatternsAll()
 

epyt_flow/metrics.py

@@ -3,12 +3,74 @@ This module provides different metrics for evaluation.
 """
 import numpy as np
 from sklearn.metrics import roc_auc_score as skelarn_roc_auc_score, f1_score as skelarn_f1_scpre, \
-    mean_absolute_error
+    mean_absolute_error, root_mean_squared_error, r2_score as sklearn_r2_score
 
 
-def running_mse(y_pred: np.ndarray, y: np.ndarray):
+def r2_score(y_pred: np.ndarray, y: np.ndarray) -> float:
     """
-    Computes the running Mean Squared Error (MSE).
+    Computes the R^2 score (also called the coefficient of determination).
+
+    Parameters
+    ----------
+    y_pred : `numpy.ndarray`
+        Predicted outputs.
+    y : `numpy.ndarray`
+        Ground truth outputs.
+
+    Returns
+    -------
+    `float`
+        R^2 score.
+    """
+    return sklearn_r2_score(y, y_pred)
+
+
+def running_r2_score(y_pred: np.ndarray, y: np.ndarray) -> list[float]:
+    """
+    Computes and returns the running R^2 score -- i.e. the R^2 score for every point in time.
+
+    Parameters
+    ----------
+    y_pred : `numpy.ndarray`
+        Predicted outputs.
+    y : `numpy.ndarray`
+        Ground truth outputs.
+
+    Returns
+    -------
+    `list[float]`
+        The running R^2 score.
+    """
+    r = []
+
+    for t in range(2, len(y_pred)):
+        r.append(r2_score(y_pred[:t], y[:t]))
+
+    return r
+
+
+def mean_squared_error(y_pred: np.ndarray, y: np.ndarray) -> float:
+    """
+    Computes the Mean Squared Error (MSE).
+
+    Parameters
+    ----------
+    y_pred : `numpy.ndarray`
+        Predicted outputs.
+    y : `numpy.ndarray`
+        Ground truth outputs.
+
+    Returns
+    -------
+    `float`
+        MSE.
+    """
+    return root_mean_squared_error(y, y_pred)**2
+
+
+def running_mse(y_pred: np.ndarray, y: np.ndarray) -> list[float]:
+    """
+    Computes the running Mean Squared Error (MSE) -- i.e. the MSE for every point in time.
 
     Parameters
     ----------
@@ -39,7 +101,7 @@ def running_mse(y_pred: np.ndarray, y: np.ndarray):
     r_mse = list(esq for esq in e_sq)
 
     for i in range(1, len(y)):
-        r_mse[i] = ((i * r_mse[i - 1]) / (i + 1)) + (r_mse[i] / (i + 1))
+        r_mse[i] = float((i * r_mse[i - 1]) / (i + 1)) + (r_mse[i] / (i + 1))
 
     return r_mse
 

epyt_flow/serialization.py

@@ -309,6 +309,39 @@ class JsonSerializable(Serializable):
         """
         return my_load_from_json(data)
 
+    @staticmethod
+    def load_from_json_file(f_in: str) -> Any:
+        """
+        Deserializes an instance of this class from a JSON file.
+
+        Parameters
+        ----------
+        f_in : `str`
+            Path to the JSON file from which to deserialize the object.
+
+        Returns
+        -------
+        `Any`
+            Deserialized object.
+        """
+        with open(f_in, "r", encoding="utf-8") as f:
+            return my_load_from_json(f.read())
+
+    def save_to_json_file(self, f_out: str) -> None:
+        """
+        Serializes this instance and stores it in a JSON file.
+
+        Parameters
+        ----------
+        f_in : `str`
+            Path to the JSON file where this serialized object will be stored.
+        """
+        if not f_out.endswith(self.file_ext()):
+            f_out += self.file_ext()
+
+        with open(f_out, "w", encoding="utf-8") as f:
+            f.write(self.to_json())
+
 
 def load(data: Union[bytes, BufferedIOBase]) -> Any:
     """