public void RMSETest()
{
/*
* 5 3 0 1
* 4 0 0 1
* 1 1 0 5
* 1 0 0 4
* 0 1 5 4
*/
DataMatrix R = GetSampleRatingMatrix();
/*
* 3 3 0 1
* 4 0 0 5
* 1 2 0 5
* 1 0 0 4
* 0 1 5 4
*/
DataMatrix R_predicted = GetSampleRatingMatrix();
R_predicted[0, 0] = 3; // was 5
R_predicted[2, 1] = 2; // was 1
R_predicted[1, 3] = 5; // was 1
double debug1 = RMSE.Evaluate(R, R_predicted);
double debug2 = Math.Sqrt(21.0 / 13.0);
Debug.Assert(debug1 == debug2);
}
public string RunNMFbasedOMF(int maxEpoch, double learnRate, double regularization, int factorCount,
List <double> quantizer, int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("NMF based OMF"));
// NMF Prediction
// Get ratings from scorer, for both train and test
// R_all contains indexes of all ratings both train and test
DataMatrix R_all = new DataMatrix(R_unknown.UserCount, R_unknown.ItemCount);
R_all.MergeNonOverlap(R_unknown);
R_all.MergeNonOverlap(R_train.IndexesOfNonZeroElements());
Utils.StartTimer();
DataMatrix R_predictedByNMF = NMF.PredictRatings(R_train, R_all, maxEpoch,
learnRate, regularization, factorCount);
log.AppendLine(Utils.StopTimer());
// OMF Prediction
log.AppendLine(Utils.PrintHeading("Ordinal Matrix Factorization with NMF as scorer"));
Utils.StartTimer();
Dictionary <Tuple <int, int>, List <double> > OMFDistributionByUserItem;
DataMatrix R_predicted;
log.AppendLine(OMF.PredictRatings(R_train.Matrix, R_unknown.Matrix, R_predictedByNMF.Matrix,
quantizer, out R_predicted, out OMFDistributionByUserItem));
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
// Save OMFDistribution to file
if (!File.Exists(GetDataFileName("RatingOMF_")))
{
Utils.IO <Dictionary <Tuple <int, int>, List <double> > > .SaveObject(OMFDistributionByUserItem, GetDataFileName("RatingOMF_"));
}
return(log.ToString());
}
public string RunNMFbasedORF(double regularization, double learnRate,
int maxEpoch, List <double> quantizer, int topN = 0)
{
// Load OMFDistribution from file
Dictionary <Tuple <int, int>, List <double> > OMFDistributionByUserItem;
if (File.Exists(GetDataFileName("RatingOMF_")))
{
OMFDistributionByUserItem = Utils.IO <Dictionary <Tuple <int, int>, List <double> > > .LoadObject(GetDataFileName("RatingOMF_"));
}
else
{
return("Abort, Run OMF first.");
}
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("NMF based ORF"));
// Prediction
Utils.StartTimer();
DataMatrix R_predicted_expectations;
DataMatrix R_predicted_mostlikely;
ORF orf = new ORF();
orf.PredictRatings(R_train, R_unknown, StrongSimilarityIndicatorsByItemRating,
OMFDistributionByUserItem, regularization, learnRate, maxEpoch,
quantizer.Count, out R_predicted_expectations, out R_predicted_mostlikely);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted_expectations).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", RMSE.Evaluate(R_test, R_predicted_mostlikely).ToString("0.0000")));
// Top-N Evaluation
if (topN != 0)
{
var topNItemsByUser_expectations = ItemRecommendationCore.GetTopNItemsByUser(R_predicted_expectations, topN);
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser_expectations, n).ToString("0.0000")));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser_expectations, n).ToString("0.0000")));
}
}
return(log.ToString());
}
public void RMSETest()
{
var actual = new Matrix(100, 1);
var expected = new Matrix(100, 1);
actual.InRandomize(0.25, 0.75);
expected.InRandomize(0.25, 0.75);
var metric = new RMSE();
var e = metric.Evaluate(actual, expected);
var val = Math.Sqrt(new MSE().Evaluate(actual, expected));
Assert.IsTrue(Math.Abs(e - val) < 0.01, metric.Type().ToString() + " Evaluate.");
}
/// <summary>
/// Rating based Non-negative Matrix Factorization
/// </summary>
public string RunNMF(int maxEpoch, double learnRate, double regularization,
int factorCount, int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("NMF"));
// Prediction
Utils.StartTimer();
DataMatrix R_predicted = NMF.PredictRatings(R_train, R_unknown, maxEpoch,
learnRate, regularization, factorCount);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
return(log.ToString());
}
/// <summary>
/// Predict all unknown values as global mean rating.
/// </summary>
public string RunGlobalMean()
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("Global Mean"));
// Prediction
Utils.StartTimer();
double globalMean = R_train.GetGlobalMean();
DataMatrix R_predicted = R_unknown.Multiply(globalMean);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
return(log.ToString());
}
public string RunUserKNN(int topN = 0)
{
if (!ReadyForNumerical)
{
GetReadyForNumerical();
}
StringBuilder log = new StringBuilder();
log.AppendLine(Utils.PrintHeading("UserKNN"));
// Prediction
Utils.StartTimer();
DataMatrix R_predicted = Numerical.UserKNN.PredictRatings(R_train, R_unknown, UserSimilaritiesOfRating, MaxCountOfNeighbors);
log.AppendLine(Utils.StopTimer());
// Numerical Evaluation
log.AppendLine(Utils.PrintValue("RMSE", RMSE.Evaluate(R_test, R_predicted).ToString("0.0000")));
log.AppendLine(Utils.PrintValue("MAE", MAE.Evaluate(R_test, R_predicted).ToString("0.0000")));
// TopN Evaluation
if (topN != 0)
{
var topNItemsByUser = ItemRecommendationCore.GetTopNItemsByUser(R_predicted, topN);
for (int n = 1; n <= topN; n++)
{
Utils.PrintValue("NCDG@" + n, NCDG.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000"));
}
for (int n = 1; n <= topN; n++)
{
log.AppendLine(Utils.PrintValue("MAP@" + n, MAP.Evaluate(RelevantItemsByUser, topNItemsByUser, n).ToString("0.0000")));
}
}
return(log.ToString());
}
