Abstract base class for all classifiers. More...

#include <classifier.h>

Inheritance diagram for cmp::classifier::Classifier:

[legend]

Public Member Functions
virtual	~Classifier ()=default

virtual std::vector< double >	predictProbabilities (const Eigen::Ref< const Eigen::VectorXd > &x) const =0

virtual size_t	predict (const Eigen::Ref< const Eigen::VectorXd > &x) const =0

Detailed Description

Abstract base class for all classifiers.

Mathematical Formulation A classifier maps an input vector \(\mathbf{x} \in \mathbb{R}^D\) to a class label \(y \in \{0, 1, \dots, C-1\}\), or estimates the class posterior probabilities \(P(y = c \mid \mathbf{x})\) for \(c = 0, \dots, C-1\).

Implementation Algorithm Provides a pure virtual interface defining:

predictProbabilities(x): Returns a vector of probability estimates.
predict(x): Returns the class label with the highest probability.

Constructor & Destructor Documentation

◆ ~Classifier()

virtual cmp::classifier::Classifier::~Classifier ( )

virtualdefault

Member Function Documentation

◆ predict()

virtual size_t cmp::classifier::Classifier::predict ( const Eigen::Ref< const Eigen::VectorXd > & x ) const

pure virtual

Implemented in cmp::classifier::KDE, cmp::classifier::SVM, and cmp::classifier::Dummy.

◆ predictProbabilities()

virtual std::vector< double > cmp::classifier::Classifier::predictProbabilities ( const Eigen::Ref< const Eigen::VectorXd > & x ) const