mlpack
3.0.0
|
This class implements Collaborative Filtering (CF). More...
Public Member Functions | |
CF (const size_t numUsersForSimilarity=5, const size_t rank=0) | |
Initialize the CF object without performing any factorization. More... | |
template < typename FactorizerType = amf::NMFALSFactorizer > | |
CF (const arma::mat &data, FactorizerType factorizer=FactorizerType(), const size_t numUsersForSimilarity=5, const size_t rank=0) | |
Initialize the CF object using an instantiated factorizer, immediately factorizing the given data to create a model. More... | |
template < typename FactorizerType = amf::NMFALSFactorizer > | |
CF (const arma::sp_mat &data, FactorizerType factorizer=FactorizerType(), const size_t numUsersForSimilarity=5, const size_t rank=0, const typename std::enable_if_t< !FactorizerTraits< FactorizerType >::UsesCoordinateList > *=0) | |
Initialize the CF object using an instantiated factorizer, immediately factorizing the given data to create a model. More... | |
const arma::sp_mat & | CleanedData () const |
Get the cleaned data matrix. More... | |
void | GetRecommendations (const size_t numRecs, arma::Mat< size_t > &recommendations) |
Generates the given number of recommendations for all users. More... | |
void | GetRecommendations (const size_t numRecs, arma::Mat< size_t > &recommendations, const arma::Col< size_t > &users) |
Generates the given number of recommendations for the specified users. More... | |
const arma::mat & | H () const |
Get the Item Matrix. More... | |
void | NumUsersForSimilarity (const size_t num) |
Sets number of users for calculating similarity. More... | |
size_t | NumUsersForSimilarity () const |
Gets number of users for calculating similarity. More... | |
double | Predict (const size_t user, const size_t item) const |
Predict the rating of an item by a particular user. More... | |
void | Predict (const arma::Mat< size_t > &combinations, arma::vec &predictions) const |
Predict ratings for each user-item combination in the given coordinate list matrix. More... | |
void | Rank (const size_t rankValue) |
Sets rank parameter for matrix factorization. More... | |
size_t | Rank () const |
Gets rank parameter for matrix factorization. More... | |
template < typename Archive > | |
void | serialize (Archive &ar, const unsigned int) |
Serialize the CF model to the given archive. More... | |
template < typename FactorizerType = amf::NMFALSFactorizer > | |
void | Train (const arma::mat &data, FactorizerType factorizer=FactorizerType()) |
Train the CF model (i.e. More... | |
template < typename FactorizerType = amf::NMFALSFactorizer > | |
void | Train (const arma::sp_mat &data, FactorizerType factorizer=FactorizerType(), const typename std::enable_if_t< !FactorizerTraits< FactorizerType >::UsesCoordinateList > *=0) |
Train the CF model (i.e. More... | |
const arma::mat & | W () const |
Get the User Matrix. More... | |
Static Public Member Functions | |
static void | CleanData (const arma::mat &data, arma::sp_mat &cleanedData) |
Converts the User, Item, Value Matrix to User-Item Table. More... | |
Detailed Description
This class implements Collaborative Filtering (CF).
This implementation presently supports Alternating Least Squares (ALS) for collaborative filtering.
A simple example of how to run Collaborative Filtering is shown below.
The data matrix is a (user, item, rating) table. Each column in the matrix should have three rows. The first represents the user; the second represents the item; and the third represents the rating. The user and item, while they are in a matrix that holds doubles, should hold integer (or size_t) values. The user and item indices are assumed to start at 0.
- Template Parameters
-
FactorizerType The type of matrix factorization to use to decompose the rating matrix (a W and H matrix). This must implement the method Apply(arma::sp_mat& data, size_t rank, arma::mat& W, arma::mat& H).
Constructor & Destructor Documentation
◆ CF() [1/3]
CF | ( | const size_t | numUsersForSimilarity = 5 , |
const size_t | rank = 0 |
||
) |
Initialize the CF object without performing any factorization.
Be sure to call Train() before calling GetRecommendations() or any other functions!
◆ CF() [2/3]
CF | ( | const arma::mat & | data, |
FactorizerType | factorizer = FactorizerType() , |
||
const size_t | numUsersForSimilarity = 5 , |
||
const size_t | rank = 0 |
||
) |
Initialize the CF object using an instantiated factorizer, immediately factorizing the given data to create a model.
There are parameters that can be set; default values are provided for each of them. If the rank is left unset (or is set to 0), a simple density-based heuristic will be used to choose a rank.
The provided dataset should be a coordinate list; that is, a 3-row matrix where each column corresponds to a (user, item, rating) entry in the matrix.
- Parameters
-
data Data matrix: coordinate list or dense matrix. factorizer Instantiated factorizer object. numUsersForSimilarity Size of the neighborhood. rank Rank parameter for matrix factorization.
◆ CF() [3/3]
CF | ( | const arma::sp_mat & | data, |
FactorizerType | factorizer = FactorizerType() , |
||
const size_t | numUsersForSimilarity = 5 , |
||
const size_t | rank = 0 , |
||
const typename std::enable_if_t< !FactorizerTraits< FactorizerType >::UsesCoordinateList > * | = 0 |
||
) |
Initialize the CF object using an instantiated factorizer, immediately factorizing the given data to create a model.
There are parameters that can be set; default values are provided for each of them. If the rank is left unset (or is set to 0), a simple density-based heuristic will be used to choose a rank. Data will be considered in the format of items vs. users and will be passed directly to the factorizer without cleaning. This overload of the constructor will only be available if the factorizer does not use a coordinate list (i.e. if UsesCoordinateList is false).
The U and T template parameters are for SFINAE, so that this overload is only available when the FactorizerType uses a coordinate list.
- Parameters
-
data Sparse matrix data. factorizer Instantiated factorizer object. numUsersForSimilarity Size of the neighborhood. rank Rank parameter for matrix factorization.
Member Function Documentation
◆ CleanData()
|
static |
Converts the User, Item, Value Matrix to User-Item Table.
◆ CleanedData()
|
inline |
◆ GetRecommendations() [1/2]
void GetRecommendations | ( | const size_t | numRecs, |
arma::Mat< size_t > & | recommendations | ||
) |
Generates the given number of recommendations for all users.
- Parameters
-
numRecs Number of Recommendations recommendations Matrix to save recommendations into.
◆ GetRecommendations() [2/2]
void GetRecommendations | ( | const size_t | numRecs, |
arma::Mat< size_t > & | recommendations, | ||
const arma::Col< size_t > & | users | ||
) |
Generates the given number of recommendations for the specified users.
- Parameters
-
numRecs Number of Recommendations recommendations Matrix to save recommendations users Users for which recommendations are to be generated
◆ H()
◆ NumUsersForSimilarity() [1/2]
|
inline |
◆ NumUsersForSimilarity() [2/2]
|
inline |
◆ Predict() [1/2]
double Predict | ( | const size_t | user, |
const size_t | item | ||
) | const |
Predict the rating of an item by a particular user.
- Parameters
-
user User to predict for. item Item to predict for.
◆ Predict() [2/2]
void Predict | ( | const arma::Mat< size_t > & | combinations, |
arma::vec & | predictions | ||
) | const |
Predict ratings for each user-item combination in the given coordinate list matrix.
The matrix 'combinations' should have two rows and number of columns equal to the number of desired predictions. The first element of each column corresponds to the user index, and the second element of each column corresponds to the item index. The output vector 'predictions' will have length equal to combinations.n_cols, and predictions[i] will be equal to the prediction for the user/item combination in combinations.col(i).
- Parameters
-
combinations User/item combinations to predict. predictions Predicted ratings for each user/item combination.
◆ Rank() [1/2]
|
inline |
◆ Rank() [2/2]
|
inline |
◆ serialize()
void serialize | ( | Archive & | ar, |
const unsigned | int | ||
) |
Serialize the CF model to the given archive.
◆ Train() [1/2]
void Train | ( | const arma::mat & | data, |
FactorizerType | factorizer = FactorizerType() |
||
) |
Train the CF model (i.e.
factorize the input matrix) using the parameters that have already been set for the model (specifically, the rank parameter), and optionally, using the given FactorizerType.
- Parameters
-
data Input dataset; coordinate list or dense matrix. factorizer Instantiated factorizer.
◆ Train() [2/2]
void Train | ( | const arma::sp_mat & | data, |
FactorizerType | factorizer = FactorizerType() , |
||
const typename std::enable_if_t< !FactorizerTraits< FactorizerType >::UsesCoordinateList > * | = 0 |
||
) |
Train the CF model (i.e.
factorize the input matrix) using the parameters that have already been set for the model (specifically, the rank parameter), and optionally, using the given FactorizerType.
- Parameters
-
data Sparse matrix data. factorizer Instantiated factorizer.
◆ W()
The documentation for this class was generated from the following file:
- src/mlpack/methods/cf/cf.hpp
Generated by
