C# Class MyMediaLite.Eval.Ratings

Evaluation class for rating prediction

Public Methods

Method	Description
DisplayResults ( double>.Dictionary result ) : void	Write rating prediction results to STDOUT
Evaluate ( IRatingPredictor recommender, IRatings ratings ) : double>.Dictionary	Evaluates a rating predictor for RMSE, MAE, and NMAE For NMAE, see "Eigentaste: A Constant Time Collaborative Filtering Algorithm" by Goldberg et al.
EvaluateOnSplit ( RatingPredictor recommender, ISplit split ) : double>.Dictionary	Evaluate on the folds of a dataset split
EvaluateOnSplit ( RatingPredictor recommender, ISplit split, bool show_results ) : double>.Dictionary	Evaluate on the folds of a dataset split
EvaluateOnline ( IRatingPredictor recommender, IRatings ratings ) : double>.Dictionary	Online evaluation for rating prediction Every rating that is tested is added to the training set afterwards.

Write rating prediction results to STDOUT

public static DisplayResults ( double>.Dictionary result ) : void
result	double>.Dictionary	the output of the Evaluate() method
return	void

Evaluates a rating predictor for RMSE, MAE, and NMAE

For NMAE, see "Eigentaste: A Constant Time Collaborative Filtering Algorithm" by Goldberg et al.

public static Evaluate ( IRatingPredictor recommender, IRatings ratings ) : double>.Dictionary
recommender	IRatingPredictor	rating predictor
ratings	IRatings	Test cases
return	double>.Dictionary

Evaluate on the folds of a dataset split

public static EvaluateOnSplit ( RatingPredictor recommender, ISplit split ) : double>.Dictionary
recommender	MyMediaLite.RatingPrediction.RatingPredictor	a rating predictor
split	ISplit	a rating dataset split
return	double>.Dictionary

Evaluate on the folds of a dataset split

public static EvaluateOnSplit ( RatingPredictor recommender, ISplit split, bool show_results ) : double>.Dictionary
recommender	MyMediaLite.RatingPrediction.RatingPredictor	a rating predictor
split	ISplit	a rating dataset split
show_results	bool	set to true to print results to STDERR
return	double>.Dictionary

Online evaluation for rating prediction

Every rating that is tested is added to the training set afterwards.

public static EvaluateOnline ( IRatingPredictor recommender, IRatings ratings ) : double>.Dictionary
recommender	IRatingPredictor	rating predictor
ratings	IRatings	Test cases
return	double>.Dictionary