slope::MAE Class Reference

Mean Absolute Error (MAE) scoring metric. More...

#include <score.h>

Inheritance diagram for slope::MAE:

Collaboration diagram for slope::MAE:

Public Member Functions
double	eval (const Eigen::MatrixXd &eta, const Eigen::MatrixXd &y, const std::unique_ptr< Loss > &) const override
Public Member Functions inherited from slope::MinimizeScore
bool	isWorse (double a, double b) const override
double	initValue () const override
Public Member Functions inherited from slope::Score
std::function< bool(double, double)>	getComparator () const

Additional Inherited Members
Static Public Member Functions inherited from slope::Score
static std::unique_ptr< Score >	create (const std::string &metric)

Detailed Description

Mean Absolute Error (MAE) scoring metric.

Computes the average absolute difference between predictions and true responses. Inherits from MinimizeScore since lower MAE values indicate better fit.

MAE = (1/n) Σ|y_i - η_i| where:

• n is the number of samples
• y_i is the true response for sample i
• η_i is the predicted response for sample i

MAE is more robust to outliers compared to MSE as it uses absolute rather than squared differences.

Definition at line 189 of file score.h.

Member Function Documentation

eval()

double slope::MAE::eval ( const Eigen::MatrixXd &  eta, const Eigen::MatrixXd &  y, const std::unique_ptr< Loss > &    ) const

overridevirtual

Evaluates the MAE between predictions and true responses.

Parameters

eta	Matrix of model predictions
y	Matrix of true responses

Returns

Computed MAE value

Implements slope::Score.

---

The documentation for this class was generated from the following file:

• include/slope/score.h