org.apache.spark.mllib.recommendation

Class MatrixFactorizationModel

Object

org.apache.spark.mllib.recommendation.MatrixFactorizationModel

All Implemented Interfaces:

java.io.Serializable

public class MatrixFactorizationModel
extends Object
implements scala.Serializable

Model representing the result of matrix factorization.

See Also:

Serialized Form

Method Summary

Methods

Modifier and Type	Method and Description
double	predict(int user, int product) Predict the rating of one user for one product.
JavaRDD<byte[]>	predict(JavaRDD<byte[]> usersProductsJRDD) :: DeveloperApi :: Predict the rating of many users for many products.
RDD<Rating>	predict(RDD<scala.Tuple2<Object,Object>> usersProducts) Predict the rating of many users for many products.
RDD<scala.Tuple2<Object,double[]>>	productFeatures()
int	rank()
RDD<scala.Tuple2<Object,double[]>>	userFeatures()

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

rank

public int rank()

userFeatures

public RDD<scala.Tuple2<Object,double[]>> userFeatures()

productFeatures

public RDD<scala.Tuple2<Object,double[]>> productFeatures()

predict

public double predict(int user,
             int product)

Predict the rating of one user for one product.

predict

public RDD<Rating> predict(RDD<scala.Tuple2<Object,Object>> usersProducts)

Predict the rating of many users for many products. The output RDD has an element per each element in the input RDD (including all duplicates) unless a user or product is missing in the training set.

Parameters:

usersProducts - RDD of (user, product) pairs.

Returns:

RDD of Ratings.

predict

public JavaRDD<byte[]> predict(JavaRDD<byte[]> usersProductsJRDD)

:: DeveloperApi :: Predict the rating of many users for many products. This is a Java stub for python predictAll()

Parameters:

usersProductsJRDD - A JavaRDD with serialized tuples (user, product)

Returns:

JavaRDD of serialized Rating objects.