SG++-Doxygen-Documentation
Loading...
Searching...
No Matches
Public Member Functions | List of all members
sgpp::datadriven::Accuracy Class Reference

Metric that quantifies the difference between predicted values and actual values in terms of mean squared error (MSE). More...

#include <Accuracy.hpp>

Inheritance diagram for sgpp::datadriven::Accuracy:
sgpp::datadriven::Metric

Public Member Functions

Metric * clone () const override
 Polymorphic clone pattern.
 
double measure (const DataVector &predictedValues, const DataVector &trueValues, const ModelFittingBase &model, Dataset &testDataset) const override
 Quantify the difference between predicted values and actual values in terms of accuracy.
 
double measureLowerIsBetter (const DataVector &predictedValues, const DataVector &trueValues, const ModelFittingBase &model, Dataset &testDataset) const override
 Quantify the difference between predicted values and actual values in terms of accuracy.
 
- Public Member Functions inherited from sgpp::datadriven::Metric
 Metric ()=default
 Default constructor.
 
 Metric (const Metric &rhs)=default
 Copy constructor.
 
 Metric (Metric &&rhs)=default
 Move constructor.
 
Metric & operator= (const Metric &rhs)=default
 Copy assign operator.
 
Metric & operator= (Metric &&rhs)=default
 Move assign operator.
 
virtual ~Metric ()=default
 virtual destructor.
 

Detailed Description

Metric that quantifies the difference between predicted values and actual values in terms of mean squared error (MSE).

MSE is defined strictly positive such that smaller values are better.

Member Function Documentation

◆ clone()

Metric * sgpp::datadriven::Accuracy::clone ( ) const
overridevirtual

Polymorphic clone pattern.

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

Implements sgpp::datadriven::Metric.

◆ measure()

double sgpp::datadriven::Accuracy::measure ( const DataVector &  predictedValues,
const DataVector &  trueValues,
const ModelFittingBase &  model,
Dataset &  testDataset 
) const
overridevirtual

Quantify the difference between predicted values and actual values in terms of accuracy.

Parameters
predictedValuesvalues calculated by the model for testing data
trueValuesactual values as taken from the dataset.
modelreference to the model
testDatasetdataset with test data
Returns
Accuracy - larger values are better.

Implements sgpp::datadriven::Metric.

References sgpp::base::DataVector::get().

◆ measureLowerIsBetter()

double sgpp::datadriven::Accuracy::measureLowerIsBetter ( const DataVector &  predictedValues,
const DataVector &  trueValues,
const ModelFittingBase &  model,
Dataset &  testDataset 
) const
overridevirtual

Quantify the difference between predicted values and actual values in terms of accuracy.

Parameters
predictedValuesvalues calculated by the model for testing data
trueValuesactual values as taken from the dataset.
modelreference to the model
testDatasetdataset with test data
Returns
inverse accuracy - smaller values are better.

Implements sgpp::datadriven::Metric.

References sgpp::datadriven::accuracy, and sgpp::base::DataVector::get().

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