MyMediaLite
3.10
|
Combining several predictors with a weighted ensemble More...
Public Member Functions | |
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 | |
Object | Clone () |
create a shallow copy of the object | |
override void | LoadModel (string file) |
Get the model parameters from a file | |
override float | Predict (int user_id, int item_id) |
Predict rating or score for a given user-item combination | |
override IList< Tuple< int, float > > | Recommend (int user_id, int n=20, ICollection< int > ignore_items=null, ICollection< int > candidate_items=null) |
Recommend items for a given user | |
override void | SaveModel (string file) |
Save the model parameters to a file | |
string | ToString () |
Return a string representation of the recommender | |
override void | Train () |
Learn the model parameters of the recommender from the training data |
Public Attributes | |
IList< IRecommender > | recommenders = new List<IRecommender>() |
list of recommenders | |
IList< float > | weights = new List<float>() |
List of component weights |
Protected Attributes | |
double | weight_sum |
Sum of the component weights |
Properties | |
float | MaxRating [get, set] |
The max rating value | |
float | MinRating [get, set] |
The min rating value |
Combining several predictors with a weighted ensemble
This recommender does NOT support incremental updates.
|
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.
user_id | the user ID |
item_id | the item ID |
Implements IRecommender.
|
inlineinherited |
create a shallow copy of the object
|
inlinevirtual |
Get the model parameters from a file
filename | the name of the file to read from |
Implements Ensemble.
|
inlinevirtual |
Predict rating or score for a given user-item combination
user_id | the user ID |
item_id | the item ID |
Implements Ensemble.
|
inlinevirtual |
Recommend items for a given user
user_id | the user ID |
n | the number of items to recommend, -1 for as many as possible |
ignore_items | collection if items that should not be returned; if null, use empty collection |
candidate_items | the candidate items to choose from; if null, use all items |
Implements Ensemble.
|
inlinevirtual |
Save the model parameters to a file
filename | the name of the file to write to |
Implements Ensemble.
|
inherited |
Return a string representation of the recommender
The ToString() method of recommenders should list the class name and all hyperparameters, separated by space characters.
Implemented in BPRMF, BiasedMatrixFactorization, SVDPlusPlus, MatrixFactorization, SigmoidCombinedAsymmetricFactorModel, CoClustering, BPRSLIM, SigmoidItemAsymmetricFactorModel, LeastSquareSLIM, TimeAwareBaseline, SigmoidUserAsymmetricFactorModel, LatentFeatureLogLinearModel, FactorWiseMatrixFactorization, UserItemBaseline, SigmoidSVDPlusPlus, SocialMF, BPRLinear, KNN, NaiveBayes, WRMF, KNN, MostPopular, TimeAwareBaselineWithFrequencies, SoftMarginRankingMF, Recommender, ExternalItemRecommender, ExternalRatingPredictor, WeightedBPRMF, MultiCoreBPRMF, and Constant.
|
inlinevirtual |
Learn the model parameters of the recommender from the training data
Reimplemented from Ensemble.
|
inherited |
list of recommenders
|
protected |
Sum of the component weights
IList<float> weights = new List<float>() |
List of component weights
|
getsetinherited |
The max rating value
The max rating value
|
getsetinherited |
The min rating value
The min rating value