Global heuristic to learn a model. More...

#include <HeuristicPipeline.h>

Public Member Functions
	HeuristicPipeline (Config &config, AlternativesPerformance &altPerfs)

MRSortModel	start ()

void	orderModels ()

void	customSort ()

void	computeAccuracy (MRSortModel &model)

Public Attributes
std::vector< MRSortModel >	models

Detailed Description

Global heuristic to learn a model.

The heuristic pipeline is responsible of initializing the sub algorithms: the WeightUpdater, the ProfileInitializer and the ProfileUpdater, is responsible for initilizing the population of n models and run the learning metaheuristic to converge into a learned model.

A complete description of the heuristic can be found in Learning Algorithms.

Constructor & Destructor Documentation

◆ HeuristicPipeline()

HeuristicPipeline::HeuristicPipeline	(	Config &	config,
		AlternativesPerformance &	altPerfs
	)

HeuristicPipeline Base constructor, stores the given config.

Parameters

config app config to start and run the application.

Member Function Documentation

◆ computeAccuracy()

void HeuristicPipeline::computeAccuracy ( MRSortModel & model )

computeAccuracy compute the accuracy of the model given the dataset and store it into the model

Parameters

model model to work on

◆ customSort()

void HeuristicPipeline::customSort ( )

customSort sort the model given their accuracy

◆ orderModels()

void HeuristicPipeline::orderModels ( )

orderModels Order models by their accuracy.

◆ start()

MRSortModel HeuristicPipeline::start ( )

Start run the heuristic pipeline and return the best model learned.

Returns: best_model

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

/Users/ThomasVindard 1/Desktop/3A/Projet/fastPL/include/learning/HeuristicPipeline.h

Public Member Functions

Public Attributes