Moral88 0.5.0__tar.gz → 0.7.0__tar.gz

{Moral88-0.5.0 → Moral88-0.7.0}/Moral88/regression.py

@@ -31,6 +31,7 @@ class DataValidator:
         """
         Returns the number of samples in the array.
         """
+        array = np.asarray(array)
         if hasattr(array, 'shape') and len(array.shape) > 0:
             return array.shape[0]
         else:
@@ -100,13 +101,17 @@ class DataValidator:
         return y_true, y_pred
 
 
-class Metrics:
-    def mean_bias_deviation(self, y_true, y_pred, library=None):
+class metrics:
+    def mean_bias_deviation(self, y_true, y_pred, library=None, flatten=True):
         """
         Computes Mean Bias Deviation (MBD).
         """
         y_true, y_pred = self.validator.validate_mae_mse_inputs(y_true, y_pred, library)
 
+        if flatten and y_true.ndim > 1:
+            y_true = y_true.flatten()
+            y_pred = y_pred.flatten()
+
         if library == 'sklearn':
             # Sklearn does not have a direct implementation for MBD
             raise NotImplementedError("Mean Bias Deviation is not implemented in sklearn.")
@@ -130,12 +135,16 @@ class Metrics:
     def __init__(self):
         self.validator = DataValidator()
 
-    def r2_score(self, y_true, y_pred, sample_weight=None, library=None):
+    def r2_score(self, y_true, y_pred, sample_weight=None, library=None, flatten=True):
         """
         Computes R2 score.
         """
         y_true, y_pred, sample_weight = self.validator.validate_r2_score_inputs(y_true, y_pred, sample_weight)
 
+        if flatten and y_true.ndim > 1:
+            y_true = y_true.flatten()
+            y_pred = y_pred.flatten()
+
         if library == 'sklearn':
             from sklearn.metrics import r2_score as sklearn_r2
             return sklearn_r2(y_true, y_pred, sample_weight=sample_weight)
@@ -166,6 +175,9 @@ class Metrics:
             if threshold is not None:
                 y_pred = np.clip(y_pred, threshold[0], threshold[1])
 
+            if y_true.ndim > 1 and flatten:
+                y_true = y_true.flatten()
+                y_pred = y_pred.flatten()
             absolute_errors = np.abs(y_true - y_pred)
 
             if method == 'mean':
@@ -177,9 +189,9 @@ class Metrics:
             else:
                 raise ValueError("Invalid method. Choose from {'mean', 'sum', 'none'}.")
 
-            if normalize and method != 'none':
-                range_y = np.ptp(y_true)
-                result = result / max(abs(range_y), 1)
+            # if normalize and method != 'none':
+            #     range_y = np.ptp(y_true)
+            #     result = result / max(abs(range_y), 1)
 
             return result
 
@@ -213,6 +225,9 @@ class Metrics:
             if threshold is not None:
                 y_pred = np.clip(y_pred, threshold[0], threshold[1])
 
+            if y_true.ndim > 1 and flatten:
+                y_true = y_true.flatten()
+                y_pred = y_pred.flatten()
             squared_errors = (y_true - y_pred) ** 2
 
             if method == 'mean':
@@ -224,9 +239,9 @@ class Metrics:
             else:
                 raise ValueError("Invalid method. Choose from {'mean', 'sum', 'none'}.")
 
-            if normalize and method != 'none':
-                range_y = np.ptp(y_true)
-                result = result / max(abs(range_y), 1)
+            # if normalize and method != 'none':
+            #     range_y = np.ptp(y_true)
+            #     result = result / max(abs(range_y), 1)
 
             return result
 
@@ -277,6 +292,7 @@ class Metrics:
         """
         y_true, y_pred = self.validator.validate_regression_targets(y_true, y_pred)
         y_true, y_pred = self.validator.validate_mae_mse_inputs(y_true, y_pred, library)
+        y_true = np.clip(y_true, 1e-8, None) 
 
         if library == 'sklearn':
             from sklearn.metrics import mean_absolute_percentage_error as sklearn_mape
@@ -294,9 +310,9 @@ class Metrics:
             y_pred_tensor = tf.convert_to_tensor(y_pred, dtype=tf.float32)
             return tf.reduce_mean(tf.abs((y_true_tensor - y_pred_tensor) / tf.clip_by_value(y_true_tensor, 1e-8, tf.float32.max))).numpy() * 100
 
-        return np.mean(np.abs((y_true - y_pred) / np.clip(y_true, 1e-8, None))) * 100
+        return np.mean(np.abs((y_true - y_pred) / np.clip(np.abs(y_true), 1e-8, None))) * 100
 
-    def explained_variance_score(self, y_true, y_pred, library=None):
+    def explained_variance_score(self, y_true, y_pred, library=None, flatten=True):
         """
         Computes Explained Variance Score.
         """
@@ -326,6 +342,58 @@ class Metrics:
         denominator = np.var(y_true)
         return 1 - numerator / denominator if denominator != 0 else 0
 
+    def adjusted_r2_score(self, y_true, y_pred, n_features, library=None, flatten=True):
+        """
+        Computes Adjusted R-Squared Score.
+        
+        Parameters:
+        y_true: array-like of shape (n_samples,)
+            Ground truth (correct) target values.
+        
+        y_pred: array-like of shape (n_samples,)
+            Estimated target values.
+        
+        n_features: int
+            Number of independent features in the model.
+        
+        library: str, optional (default=None)
+            Library to use for computation. Supports {'sklearn', 'statsmodels', None}.
+        
+        flatten: bool, optional (default=True)
+            If True, flattens multidimensional arrays before computation.
+        """
+        # Validate inputs
+        y_true, y_pred, _ = self.validator.validate_r2_score_inputs(y_true, y_pred)
+
+        # Ensure inputs are 1D arrays
+        if y_true.ndim == 0 or y_pred.ndim == 0:
+            y_true = np.array([y_true])
+            y_pred = np.array([y_pred])
+
+        if flatten and y_true.ndim > 1:
+            y_true = y_true.flatten()
+            y_pred = y_pred.flatten()
+
+        if library == 'sklearn':
+            from sklearn.metrics import r2_score
+            r2 = r2_score(y_true, y_pred)
+        elif library == 'statsmodels':
+            import statsmodels.api as sm
+            X = sm.add_constant(y_pred)
+            model = sm.OLS(y_true, X).fit()
+            r2 = model.rsquared
+        else:
+            numerator = np.sum((y_true - y_pred) ** 2)
+            denominator = np.sum((y_true - np.mean(y_true)) ** 2)
+            r2 = 1 - (numerator / denominator) if denominator != 0 else 0.0
+
+        n_samples = len(y_true)
+        if n_samples <= n_features + 1:
+            raise ValueError("Number of samples must be greater than number of features plus one for adjusted R-squared computation.")
+
+        adjusted_r2 = 1 - (1 - r2) * (n_samples - 1) / (n_samples - n_features - 1)
+        return adjusted_r2
+
 if __name__ == '__main__':
     # Example usage
     validator = DataValidator()
@@ -336,11 +404,16 @@ if __name__ == '__main__':
     print("1D array:", validator.is_1d_array(arr))
     print("Samples:", validator.check_samples(arr))
 
-    # Test R2 score
+    # Test MAE, MSE, R2, MBD, EV, MAPE, RMSE
     y_true = [3, -0.5, 2, 7]
     y_pred = [2.5, 0.0, 2, 8]
-    print("R2 Score:", metrics.r2_score(y_true, y_pred))
 
-    # Test MAE and MSE
     print("Mean Absolute Error:", metrics.mean_absolute_error(y_true, y_pred))
     print("Mean Squared Error:", metrics.mean_squared_error(y_true, y_pred))
+    print("R2 Score:", metrics.r2_score(y_true, y_pred))
+    print("Mean Bias Deviation: ", metrics.mean_bias_deviation(y_true, y_pred))
+    print("Explained Variance Score: ", metrics.explained_variance_score(y_true, y_pred))
+    print("Mean Absolute Percentage Error: ", metrics.mean_absolute_percentage_error(y_true, y_pred))
+    print("Root Mean Squared Error: ", metrics.root_mean_squared_error(y_true, y_pred))
+    print("adjusted_r2_score: ", metrics.adjusted_r2_score(y_true, y_pred, 2))
+

{Moral88-0.5.0 → Moral88-0.7.0/Moral88.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: Moral88
-Version: 0.5.0
+Version: 0.7.0
 Summary: A library for regression evaluation metrics.
 Author: Morteza Alizadeh
 Author-email: alizadeh.c2m@gmail.com

{Moral88-0.5.0/Moral88.egg-info → Moral88-0.7.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: Moral88
-Version: 0.5.0
+Version: 0.7.0
 Summary: A library for regression evaluation metrics.
 Author: Morteza Alizadeh
 Author-email: alizadeh.c2m@gmail.com

{Moral88-0.5.0 → Moral88-0.7.0}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name='Moral88',
-    version='0.5.0',
+    version='0.7.0',
     description='A library for regression evaluation metrics.',
     author='Morteza Alizadeh',
     author_email='alizadeh.c2m@gmail.com',

Moral88-0.5.0/Moral88/__init__.py

@@ -1 +0,0 @@
-from .regression import mean_absolute_error, mean_squared_error, r_squared