Added Mean Squared Logarithmic Error (MSLE) Loss Function (#10637)

* Added Mean Squared Logarithmic Error (MSLE)

* Added Mean Squared Logarithmic Error (MSLE)

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
This commit is contained in:
Ankit Avinash 2023-10-19 20:16:39 +05:30 committed by GitHub
parent 289a4dd6d3
commit bd3072b845
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -0,0 +1,55 @@
"""
Mean Squared Logarithmic Error (MSLE) Loss Function
Description:
MSLE measures the mean squared logarithmic difference between
true values and predicted values, particularly useful when
dealing with regression problems involving skewed or large-value
targets. It is often used when the relative differences between
predicted and true values are more important than absolute
differences.
Formula:
MSLE = (1/n) * Σ(log(1 + y_true) - log(1 + y_pred))^2
Source:
(https://insideaiml.com/blog/MeanSquared-Logarithmic-Error-Loss-1035)
"""
import numpy as np
def mean_squared_logarithmic_error(y_true: np.ndarray, y_pred: np.ndarray) -> float:
"""
Calculate the Mean Squared Logarithmic Error (MSLE) between two arrays.
Parameters:
- y_true: The true values (ground truth).
- y_pred: The predicted values.
Returns:
- msle: The Mean Squared Logarithmic Error between y_true and y_pred.
Example usage:
>>> true_values = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
>>> predicted_values = np.array([0.8, 2.1, 2.9, 4.2, 5.2])
>>> mean_squared_logarithmic_error(true_values, predicted_values)
0.0030860877925181344
>>> true_labels = np.array([1.0, 2.0, 3.0, 4.0, 5.0])
>>> predicted_probs = np.array([0.3, 0.8, 0.9, 0.2])
>>> mean_squared_logarithmic_error(true_labels, predicted_probs)
Traceback (most recent call last):
...
ValueError: Input arrays must have the same length.
"""
if len(y_true) != len(y_pred):
raise ValueError("Input arrays must have the same length.")
squared_logarithmic_errors = (np.log1p(y_true) - np.log1p(y_pred)) ** 2
return np.mean(squared_logarithmic_errors)
if __name__ == "__main__":
import doctest
doctest.testmod()