Add RMSE and Log-Cosh loss functions to loss_functions.py

Author: TASMAYU

machine_learning/loss_functions.py

@@ -667,3 +667,27 @@ def kullback_leibler_divergence(y_true: np.ndarray, y_pred: np.ndarray) -> float
     import doctest
 
     doctest.testmod()
+
+
+def root_mean_squared_error(y_true: np.ndarray, y_pred: np.ndarray) -> float:
+    """
+    Calculate the Root Mean Squared Error (RMSE) between ground truth and predicted values.
+    # ... docstring continues ...
+    """
+    # LINE 1: Check if input arrays have same length
+    if len(y_true) != len(y_pred):
+        raise ValueError("Input arrays must have the same length.")
+
+    # LINE 2: Calculate squared differences between true and predicted values
+    # (y_true - y_pred) gives errors, then we square each error
+    squared_errors = (y_true - y_pred) ** 2
+
+    # LINE 3: Calculate mean of all squared errors
+    # This gives Mean Squared Error (MSE)
+    mean_squared_error = np.mean(squared_errors)
+
+    # LINE 4: Take square root of MSE to get RMSE
+    # This brings units back to original scale
+    return np.sqrt(mean_squared_error)
+
+