sgpp::datadriven::Metric Class Reference

We use metrics to quantify approximation quality of a trained model. More...

#include <Metric.hpp>

Inheritance diagram for sgpp::datadriven::Metric:

Public Member Functions
virtual Metric *	clone () const =0
	Polymorphic clone pattern. More...
virtual double	measure (const DataVector &predictedValues, const DataVector &trueValues) const =0
	Quantify the difference between predicted values and actual values. More...
	Metric ()=default
	Default constructor. More...
	Metric (const Metric &rhs)=default
	Copy constructor. More...
	Metric (Metric &&rhs)=default
	Move constructor. More...
Metric &	operator= (const Metric &rhs)=default
	Copy assign operator. More...
Metric &	operator= (Metric &&rhs)=default
	Move assign operator. More...
virtual	~Metric ()=default
	virtual destructor. More...

Detailed Description

We use metrics to quantify approximation quality of a trained model.

It should be calculated in the way that smaller is always better. This means, for example that classification accuracy should be negated.

Constructor & Destructor Documentation

Metric() [1/3]

sgpp::datadriven::Metric::Metric ( )

default

Default constructor.

Metric() [2/3]

sgpp::datadriven::Metric::Metric ( const Metric &  rhs )

default

Copy constructor.

Parameters

rhs	const reference to the scorer object to copy from.

Metric() [3/3]

sgpp::datadriven::Metric::Metric ( Metric &&  rhs )

default

Move constructor.

Parameters

rhs	R-value reference to a scorer object to moved from.

~Metric()

virtual sgpp::datadriven::Metric::~Metric ( )

virtualdefault

virtual destructor.

Member Function Documentation

clone()

virtual Metric* sgpp::datadriven::Metric::clone ( ) const

pure virtual

Polymorphic clone pattern.

Returns

deep copy of this object. New object is owned by caller.

Implemented in sgpp::datadriven::NegativeLogLikelihood, sgpp::datadriven::Accuracy, and sgpp::datadriven::MSE.

measure()

virtual double sgpp::datadriven::Metric::measure ( const DataVector &  predictedValues, const DataVector &  trueValues  ) const

pure virtual

Quantify the difference between predicted values and actual values.

Does not have an inner state.

Parameters

predictedValues	values calculated by the model for testing data
trueValues	actual values as taken from the dataset.

Returns

Quantification of the difference. Smaller is better.

Implemented in sgpp::datadriven::NegativeLogLikelihood, sgpp::datadriven::Accuracy, and sgpp::datadriven::MSE.

operator=() [1/2]

Metric& sgpp::datadriven::Metric::operator= ( const Metric &  rhs )

default

Copy assign operator.

Parameters

rhs	const reference to the scorer object to copy from.

Returns

rerefernce to this with updated values.

operator=() [2/2]

Metric& sgpp::datadriven::Metric::operator= ( Metric &&  rhs )

default

Move assign operator.

Parameters

rhs	R-value reference to an a scorer object to move from.

Returns

rerefernce to this with updated values.

---

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

• datadriven/src/sgpp/datadriven/datamining/modules/scoring/Metric.hpp