MyMediaLite  3.11
Public Member Functions | Protected Member Functions | Protected Attributes | Properties | List of all members
MatrixFactorization Class Reference

Simple matrix factorization class, learning is performed by stochastic gradient descent (SGD) More...

Inheritance diagram for MatrixFactorization:
IncrementalRatingPredictor IIterativeModel IFoldInRatingPredictor RatingPredictor IIncrementalRatingPredictor IRatingPredictor Recommender IRatingPredictor IRatingPredictor IIncrementalRecommender IRecommender IRecommender IRecommender IRecommender BiasedMatrixFactorization SVDPlusPlus SigmoidCombinedAsymmetricFactorModel SigmoidItemAsymmetricFactorModel SigmoidUserAsymmetricFactorModel SocialMF GSVDPlusPlus SigmoidSVDPlusPlus

Public Member Functions

override void AddRatings (IRatings ratings)
 Add new ratings and perform incremental training More...
 
virtual bool CanPredict (int user_id, int item_id)
 Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination More...
 
Object Clone ()
 create a shallow copy of the object More...
 
virtual float ComputeObjective ()
 Compute the regularized loss More...
 
virtual void Iterate ()
 Run one iteration (= pass over the training data) More...
 
override void LoadModel (string filename)
 Get the model parameters from a file More...
 
 MatrixFactorization ()
 Default constructor More...
 
override float Predict (int user_id, int item_id)
 Predict the rating of a given user for a given item More...
 
IList< Tuple< int, float > > Recommend (int user_id, int n=-1, ICollection< int > ignore_items=null, ICollection< int > candidate_items=null)
 Recommend items for a given user More...
 
virtual System.Collections.Generic.IList< Tuple< int, float > > Recommend (int user_id, int n=-1, System.Collections.Generic.ICollection< int > ignore_items=null, System.Collections.Generic.ICollection< int > candidate_items=null)
 
override void RemoveItem (int item_id)
 Remove all feedback by one item More...
 
override void RemoveRatings (IDataSet ratings)
 Remove existing ratings and perform "incremental" training More...
 
override void RemoveUser (int user_id)
 Remove all feedback by one user More...
 
virtual void RetrainItem (int item_id)
 Updates the latent factors of an item More...
 
virtual void RetrainUser (int user_id)
 Updates the latent factors on a user More...
 
override void SaveModel (string filename)
 Save the model parameters to a file More...
 
IList< Tuple< int, float > > ScoreItems (IList< Tuple< int, float >> rated_items, IList< int > candidate_items)
 Rate a list of items given a list of ratings that represent a new user More...
 
override string ToString ()
 Return a string representation of the recommender More...
 
override void Train ()
 Learn the model parameters of the recommender from the training data More...
 
override void UpdateRatings (IRatings ratings)
 Update existing ratings and perform incremental training More...
 

Protected Member Functions

override void AddItem (int item_id)
 
override void AddUser (int user_id)
 
virtual float[] FoldIn (IList< Tuple< int, float >> rated_items)
 Compute parameters (latent factors) for a user represented by ratings More...
 
virtual void Iterate (IList< int > rating_indices, bool update_user, bool update_item)
 Iterate once over rating data and adjust corresponding factors (stochastic gradient descent) More...
 
float Predict (int user_id, int item_id, bool bound)
 
virtual float Predict (float[] user_vector, int item_id)
 Predict rating for a fold-in user and an item More...
 
virtual void UpdateLearnRate ()
 Updates current_learnrate after each epoch More...
 

Protected Attributes

float global_bias
 The bias (global average) More...
 
float max_rating
 Maximum rating value More...
 
float min_rating
 Minimum rating value More...
 
IRatings ratings
 rating data More...
 

Properties

float Decay [get, set]
 Multiplicative learn rate decay More...
 
double InitMean [get, set]
 Mean of the normal distribution used to initialize the factors More...
 
double InitStdDev [get, set]
 Standard deviation of the normal distribution used to initialize the factors More...
 
float LearnRate [get, set]
 Learn rate (update step size) More...
 
int MaxItemID [get, set]
 Maximum item ID More...
 
virtual float MaxRating [get, set]
 Maximum rating value More...
 
int MaxUserID [get, set]
 Maximum user ID More...
 
virtual float MinRating [get, set]
 Minimum rating value More...
 
uint NumFactors [get, set]
 Number of latent factors More...
 
uint NumIter [get, set]
 Number of iterations over the training data More...
 
virtual IRatings Ratings [get, set]
 The rating data More...
 
virtual float Regularization [get, set]
 Regularization parameter More...
 
bool UpdateItems [get, set]
 
bool UpdateUsers [get, set]
 

Detailed Description

Simple matrix factorization class, learning is performed by stochastic gradient descent (SGD)

Factorizing the observed rating values using a factor matrix for users and one for items.

NaN values in the model occur if values become too large or too small to be represented by the type float. If you encounter such problems, there are three ways to fix them: (1) (preferred) Use BiasedMatrixFactorization, which is more stable. (2) Change the range of rating values (1 to 5 works generally well with the default settings). (3) Decrease the learn_rate.

This recommender supports incremental updates.

Constructor & Destructor Documentation

MatrixFactorization ( )
inline

Default constructor

Member Function Documentation

override void AddRatings ( IRatings  ratings)
inlinevirtual

Add new ratings and perform incremental training

Parameters
ratingsthe ratings

Reimplemented from IncrementalRatingPredictor.

virtual bool CanPredict ( int  user_id,
int  item_id 
)
inlinevirtualinherited

Check whether a useful prediction (i.e. not using a fallback/default answer) can be made for a given user-item combination

It is up to the recommender implementor to decide when a prediction is useful, and to document it accordingly.

Parameters
user_idthe user ID
item_idthe item ID
Returns
true if a useful prediction can be made, false otherwise

Implements IRecommender.

Reimplemented in ExternalItemRecommender, ExternalRatingPredictor, BiPolarSlopeOne, SlopeOne, Constant, GlobalAverage, UserAverage, ItemAverage, and Random.

Object Clone ( )
inlineinherited

create a shallow copy of the object

virtual float ComputeObjective ( )
inlinevirtual
virtual float [] FoldIn ( IList< Tuple< int, float >>  rated_items)
inlineprotectedvirtual

Compute parameters (latent factors) for a user represented by ratings

Returns
a vector of latent factors
Parameters
rated_itemsa list of (item ID, rating value) pairs

Reimplemented in BiasedMatrixFactorization, SVDPlusPlus, SigmoidCombinedAsymmetricFactorModel, SigmoidUserAsymmetricFactorModel, SigmoidItemAsymmetricFactorModel, GSVDPlusPlus, and SigmoidSVDPlusPlus.

virtual void Iterate ( )
inlinevirtual

Run one iteration (= pass over the training data)

Implements IIterativeModel.

Reimplemented in BiasedMatrixFactorization.

virtual void Iterate ( IList< int >  rating_indices,
bool  update_user,
bool  update_item 
)
inlineprotectedvirtual

Iterate once over rating data and adjust corresponding factors (stochastic gradient descent)

Parameters
rating_indicesa list of indices pointing to the ratings to iterate over
update_usertrue if user factors to be updated
update_itemtrue if item factors to be updated

Reimplemented in BiasedMatrixFactorization, SVDPlusPlus, SigmoidSVDPlusPlus, SigmoidCombinedAsymmetricFactorModel, GSVDPlusPlus, SigmoidItemAsymmetricFactorModel, SigmoidUserAsymmetricFactorModel, and SocialMF.

override void LoadModel ( string  filename)
inline

Get the model parameters from a file

Parameters
filenamethe name of the file to read from

Implements IRecommender.

virtual float Predict ( float[]  user_vector,
int  item_id 
)
inlineprotectedvirtual

Predict rating for a fold-in user and an item

Parameters
user_vectora float vector representing the user
item_idthe item ID
Returns
the predicted rating

Reimplemented in SVDPlusPlus, BiasedMatrixFactorization, and SigmoidCombinedAsymmetricFactorModel.

override float Predict ( int  user_id,
int  item_id 
)
inline

Predict the rating of a given user for a given item

If the user or the item are not known to the recommender, the global average is returned. To avoid this behavior for unknown entities, use CanPredict() to check before.

Parameters
user_idthe user ID
item_idthe item ID
Returns
the predicted rating

Implements IRecommender.

IList<Tuple<int, float> > Recommend ( int  user_id,
int  n = -1,
ICollection< int >  ignore_items = null,
ICollection< int >  candidate_items = null 
)
inherited

Recommend items for a given user

Parameters
user_idthe user ID
nthe number of items to recommend, -1 for as many as possible
ignore_itemscollection if items that should not be returned; if null, use empty collection
candidate_itemsthe candidate items to choose from; if null, use all items
Returns
a sorted list of (item_id, score) tuples

Implemented in WeightedEnsemble, and Ensemble.

override void RemoveItem ( int  item_id)
inlinevirtual

Remove all feedback by one item

Parameters
item_idthe item ID

Reimplemented from IncrementalRatingPredictor.

override void RemoveRatings ( IDataSet  ratings)
inlinevirtual

Remove existing ratings and perform "incremental" training

Parameters
ratingsthe user and item IDs of the ratings to be removed

Reimplemented from IncrementalRatingPredictor.

override void RemoveUser ( int  user_id)
inlinevirtual

Remove all feedback by one user

Parameters
user_idthe user ID

Reimplemented from IncrementalRatingPredictor.

virtual void RetrainItem ( int  item_id)
inlinevirtual

Updates the latent factors of an item

Parameters
item_idthe item ID

Reimplemented in BiasedMatrixFactorization.

virtual void RetrainUser ( int  user_id)
inlinevirtual

Updates the latent factors on a user

Parameters
user_idthe user ID

Reimplemented in BiasedMatrixFactorization, and SVDPlusPlus.

override void SaveModel ( string  filename)
inline

Save the model parameters to a file

Parameters
filenamethe name of the file to write to

Implements IRecommender.

IList<Tuple<int, float> > ScoreItems ( IList< Tuple< int, float >>  rated_items,
IList< int >  candidate_items 
)
inline

Rate a list of items given a list of ratings that represent a new user

Returns
a list of int and float pairs, representing item IDs and predicted ratings
Parameters
rated_itemsthe ratings (item IDs and rating values) representing the new user
candidate_itemsthe items to be rated

Implements IFoldInRatingPredictor.

override string ToString ( )
inline

Return a string representation of the recommender

The ToString() method of recommenders should list the class name and all hyperparameters, separated by space characters.

Implements IRecommender.

override void Train ( )
inline

Learn the model parameters of the recommender from the training data

Implements IRecommender.

virtual void UpdateLearnRate ( )
inlineprotectedvirtual

Updates current_learnrate after each epoch

Reimplemented in BiasedMatrixFactorization.

override void UpdateRatings ( IRatings  ratings)
inlinevirtual

Update existing ratings and perform incremental training

Parameters
ratingsthe ratings

Reimplemented from IncrementalRatingPredictor.

Member Data Documentation

float global_bias
protected

The bias (global average)

float max_rating
protectedinherited

Maximum rating value

float min_rating
protectedinherited

Minimum rating value

IRatings ratings
protectedinherited

rating data

Property Documentation

float Decay
getset

Multiplicative learn rate decay

Applied after each epoch (= pass over the whole dataset)

double InitMean
getset

Mean of the normal distribution used to initialize the factors

double InitStdDev
getset

Standard deviation of the normal distribution used to initialize the factors

float LearnRate
getset

Learn rate (update step size)

int MaxItemID
getsetinherited

Maximum item ID

virtual float MaxRating
getsetinherited

Maximum rating value

int MaxUserID
getsetinherited

Maximum user ID

virtual float MinRating
getsetinherited

Minimum rating value

uint NumFactors
getset

Number of latent factors

uint NumIter
getset

Number of iterations over the training data

virtual IRatings Ratings
getsetinherited

The rating data

virtual float Regularization
getset

Regularization parameter


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