public void testRecommenderWithSyntheticData()
{
setUpSyntheticData();
factorizer = new ParallelSGDFactorizer(dataModel, rank, lambda, numIterations, 0.01, 1, 0, 0);
svdRecommender = new SVDRecommender(dataModel, factorizer);
/// a hold out test would be better, but this is just a toy example so we only check that the
/// factorization is close to the original matrix
IRunningAverage avg = new FullRunningAverage();
var userIDs = dataModel.GetUserIDs();
while (userIDs.MoveNext())
{
long userID = userIDs.Current;
foreach (IPreference pref in dataModel.GetPreferencesFromUser(userID))
{
double rating = pref.GetValue();
double estimate = svdRecommender.EstimatePreference(userID, pref.GetItemID());
double err = rating - estimate;
avg.AddDatum(err * err);
}
}
double rmse = Math.Sqrt(avg.GetAverage());
logger.Info("rmse: " + rmse);
Assert.True(rmse < 0.2);
}
public CachedFactorizer(ISieve sieve, IFactorizer factorizer)
{
Sieve = sieve;
Factorizer = factorizer;
_memory.Add(2);
_memory.Add(3);
_memory.Add(5);
_memory.Add(7);
_fullySeached = 7;
}
/// Create an SVDRecommender using a persistent store to cache factorizations. A factorization is loaded from the
/// store if present, otherwise a new factorization is computed and saved in the store.
///
/// The {@link #refresh(java.util.Collection) refresh} method recomputes the factorization and overwrites the store.
///
/// @param dataModel
/// @param factorizer
/// @param candidateItemsStrategy
/// @param persistenceStrategy
///
/// @throws TasteException
public SVDRecommender(IDataModel dataModel, IFactorizer factorizer, ICandidateItemsStrategy candidateItemsStrategy,
IPersistenceStrategy persistenceStrategy) : base(dataModel, candidateItemsStrategy)
{
this.factorizer = factorizer; //Preconditions.checkNotNull(factorizer);
this.persistenceStrategy = persistenceStrategy; // Preconditions.checkNotNull(persistenceStrategy);
try {
factorization = persistenceStrategy.Load();
} catch (IOException e) {
throw new TasteException("Error loading factorization", e);
}
if (factorization == null)
{
train();
}
refreshHelper = new RefreshHelper(() => {
train();
});
refreshHelper.AddDependency(GetDataModel());
refreshHelper.AddDependency(factorizer);
refreshHelper.AddDependency(candidateItemsStrategy);
}
/// Create an SVDRecommender using a persistent store to cache factorizations. A factorization is loaded from the
/// store if present, otherwise a new factorization is computed and saved in the store.
///
/// The {@link #refresh(java.util.Collection) refresh} method recomputes the factorization and overwrites the store.
///
/// @param dataModel
/// @param factorizer
/// @param persistenceStrategy
/// @throws TasteException
/// @throws IOException
public SVDRecommender(IDataModel dataModel, IFactorizer factorizer, IPersistenceStrategy persistenceStrategy)
: this(dataModel, factorizer, GetDefaultCandidateItemsStrategy(), persistenceStrategy)
{
}
public SVDRecommender(IDataModel dataModel, IFactorizer factorizer, ICandidateItemsStrategy candidateItemsStrategy) :
this(dataModel, factorizer, candidateItemsStrategy, getDefaultPersistenceStrategy())
{
}
public SVDRecommender(IDataModel dataModel, IFactorizer factorizer) :
this(dataModel, factorizer, new AllUnknownItemsCandidateItemsStrategy(), getDefaultPersistenceStrategy())
{
}
public void testRecommenderWithToyData()
{
setUpToyData();
factorizer = new ParallelSGDFactorizer(dataModel, rank, lambda, numIterations, 0.01, 1, 0,0);
svdRecommender = new SVDRecommender(dataModel, factorizer);
/// a hold out test would be better, but this is just a toy example so we only check that the
/// factorization is close to the original matrix
IRunningAverage avg = new FullRunningAverage();
var userIDs = dataModel.GetUserIDs();
while (userIDs.MoveNext()) {
long userID = userIDs.Current;
foreach (IPreference pref in dataModel.GetPreferencesFromUser(userID)) {
double rating = pref.GetValue();
double estimate = svdRecommender.EstimatePreference(userID, pref.GetItemID());
double err = rating - estimate;
avg.AddDatum(err * err);
}
}
double rmse = Math.Sqrt(avg.GetAverage());
logger.Info("rmse: " + rmse);
Assert.True(rmse < 0.2);
}
public void testFactorizerWithWithSyntheticData()
{
setUpSyntheticData();
var stopWatch = new System.Diagnostics.Stopwatch();
stopWatch.Start();
factorizer = new ParallelSGDFactorizer(dataModel, rank, lambda, numIterations, 0.01, 1, 0, 0);
Factorization factorization = factorizer.Factorize();
stopWatch.Stop();
long duration = stopWatch.ElapsedMilliseconds;
/// a hold out test would be better, but this is just a toy example so we only check that the
/// factorization is close to the original matrix
IRunningAverage avg = new FullRunningAverage();
var userIDs = dataModel.GetUserIDs();
IEnumerator<long> itemIDs;
while (userIDs.MoveNext()) {
long userID = userIDs.Current;
foreach (IPreference pref in dataModel.GetPreferencesFromUser(userID)) {
double rating = pref.GetValue();
var userVector = factorization.getUserFeatures(userID);
var itemVector = factorization.getItemFeatures(pref.GetItemID());
double estimate = vectorDot( userVector, itemVector);
double err = rating - estimate;
avg.AddDatum(err * err);
}
}
double sum = 0.0;
userIDs = dataModel.GetUserIDs();
while (userIDs.MoveNext()) {
long userID = userIDs.Current;
var userVector = factorization.getUserFeatures(userID);
double regularization = vectorDot( userVector, userVector);
sum += regularization;
}
itemIDs = dataModel.GetItemIDs();
while (itemIDs.MoveNext()) {
long itemID = itemIDs.Current;
var itemVector = factorization.getUserFeatures(itemID);
double regularization = vectorDot( itemVector, itemVector);
sum += regularization;
}
double rmse = Math.Sqrt(avg.GetAverage());
double loss = avg.GetAverage() / 2 + lambda / 2 * sum;
logger.Info("RMSE: " + rmse + ";\tLoss: " + loss + ";\tTime Used: " + duration + "ms");
Assert.True(rmse < 0.2);
}
public void testFactorizerWithWithSyntheticData()
{
setUpSyntheticData();
var stopWatch = new System.Diagnostics.Stopwatch();
stopWatch.Start();
factorizer = new ParallelSGDFactorizer(dataModel, rank, lambda, numIterations, 0.01, 1, 0, 0);
Factorization factorization = factorizer.Factorize();
stopWatch.Stop();
long duration = stopWatch.ElapsedMilliseconds;
/// a hold out test would be better, but this is just a toy example so we only check that the
/// factorization is close to the original matrix
IRunningAverage avg = new FullRunningAverage();
var userIDs = dataModel.GetUserIDs();
IEnumerator <long> itemIDs;
while (userIDs.MoveNext())
{
long userID = userIDs.Current;
foreach (IPreference pref in dataModel.GetPreferencesFromUser(userID))
{
double rating = pref.GetValue();
var userVector = factorization.getUserFeatures(userID);
var itemVector = factorization.getItemFeatures(pref.GetItemID());
double estimate = vectorDot(userVector, itemVector);
double err = rating - estimate;
avg.AddDatum(err * err);
}
}
double sum = 0.0;
userIDs = dataModel.GetUserIDs();
while (userIDs.MoveNext())
{
long userID = userIDs.Current;
var userVector = factorization.getUserFeatures(userID);
double regularization = vectorDot(userVector, userVector);
sum += regularization;
}
itemIDs = dataModel.GetItemIDs();
while (itemIDs.MoveNext())
{
long itemID = itemIDs.Current;
var itemVector = factorization.getUserFeatures(itemID);
double regularization = vectorDot(itemVector, itemVector);
sum += regularization;
}
double rmse = Math.Sqrt(avg.GetAverage());
double loss = avg.GetAverage() / 2 + lambda / 2 * sum;
logger.Info("RMSE: " + rmse + ";\tLoss: " + loss + ";\tTime Used: " + duration + "ms");
Assert.True(rmse < 0.2);
}
public FactorizationControl()
{
InitializeComponent();
QS.FactorizationInfo = Primes.Resources.lang.WpfControls.Factorization.Factorization.fac_result;
QS.factors = "";
QS.lblInput = "";
QS.FreeText = "100";
QS.CalcFactorText = "2";
QS.CalcBaseText = "13";
QS.CalcExpText = "2";
QS.CalcSumText = "7";
BF.FactorizationInfo = Primes.Resources.lang.WpfControls.Factorization.Factorization.fac_result;
BF.factors = "";
BF.lblInput = "";
BF.FreeText = "100";
BF.CalcFactorText = "2";
BF.CalcBaseText = "13";
BF.CalcExpText = "2";
BF.CalcSumText = "7";
_bruteforce = graph;
_bruteforce.Start += OnFactorizationStart;
_bruteforce.Stop += OnFactorizationStop_BF;
_bruteforce.FoundFactor += OnFoundFactor;
_bruteforce.Cancel += new VoidDelegate(_bruteforce_Cancel);
_rho = rhoctrl;
_rho.Start += OnFactorizationStart;
_rho.Stop += OnFactorizationStop;
_rho.FoundFactor += OnFoundFactor;
_rho.Cancel += new VoidDelegate(_rho_Cancel);
_qs = qsctrl;
_qs.Start += OnFactorizationStart;
_qs.Stop += OnFactorizationStop_QS;
_qs.FoundFactor += OnFoundFactor;
_qs.Cancel += new VoidDelegate(_qs_Cancel);
_rho.ForceGetInteger += new CallbackDelegateGetInteger(_rho_ForceGetValue);
_bruteforce.ForceGetInteger += new CallbackDelegateGetInteger(_rho_ForceGetValue);
inputnumbermanager.Execute += new ExecuteSingleDelegate(InputSingleNumberControl_Execute);
inputnumbermanager.Cancel += new VoidDelegate(InputSingleNumberControl_Cancel);
inputnumbermanager.HelpActionGenerateRandomNumber = Primes.OnlineHelp.OnlineHelpActions.Factorization_Generate;
inputnumbermanager.generateNumberControlVertFree.OnRandomNumberGenerated += new GmpBigIntegerParameterDelegate(ResetMessages);
inputnumbermanager.KeyDown += new ExecuteSingleDelegate(inputnumbermanager_ValueChanged);
SetInputValidators();
m_TbFactorizationCopy = new TextBox();
m_TbFactorizationCopy.IsReadOnly = true;
m_TbFactorizationCopy.MouseLeave += new MouseEventHandler(m_TbFactorizationCopy_MouseLeave);
m_TbFactorizationCopy.MouseMove += new MouseEventHandler(m_TbFactorizationCopy_MouseMove);
ContextMenu tbFactorizationCopyContextMenu = new ContextMenu();
MenuItem tbFactorizationCopyContextMenuCopy = new MenuItem();
tbFactorizationCopyContextMenuCopy.Click += new RoutedEventHandler(tbFactorizationCopyContextMenuCopy_Click);
tbFactorizationCopyContextMenuCopy.Header = Primes.Resources.lang.WpfControls.Factorization.Factorization.qs_copytoclipboard;
tbFactorizationCopyContextMenu.Items.Add(tbFactorizationCopyContextMenuCopy);
m_TbFactorizationCopy.ContextMenu = tbFactorizationCopyContextMenu;
UpdateMessages();
}
