Implementation of Biased Matrix Factorization on Spark

ALS2.scala

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.spark.mllib.recommendation

import org.apache.spark.Logging
import org.apache.spark.annotation.{DeveloperApi, Since}
import org.apache.spark.api.java.JavaRDD
import org.apache.spark.ml.recommendation.{ALS2 => NewALS2}
import org.apache.spark.rdd.RDD
import org.apache.spark.storage.StorageLevel

/**
 * Alternating Least Squares matrix factorization.
 *
 * ALS attempts to estimate the ratings matrix `R` as the product of two lower-rank matrices,
 * `X` and `Y`, i.e. `X * Yt = R`. Typically these approximations are called 'factor' matrices.
 * The general approach is iterative. During each iteration, one of the factor matrices is held
 * constant, while the other is solved for using least squares. The newly-solved factor matrix is
 * then held constant while solving for the other factor matrix.
 *
 * This is a blocked implementation of the ALS factorization algorithm that groups the two sets
 * of factors (referred to as "users" and "products") into blocks and reduces communication by only
 * sending one copy of each user vector to each product block on each iteration, and only for the
 * product blocks that need that user's feature vector. This is achieved by precomputing some
 * information about the ratings matrix to determine the "out-links" of each user (which blocks of
 * products it will contribute to) and "in-link" information for each product (which of the feature
 * vectors it receives from each user block it will depend on). This allows us to send only an
 * array of feature vectors between each user block and product block, and have the product block
 * find the users' ratings and update the products based on these messages.
 *
 * For implicit preference data, the algorithm used is based on
 * "Collaborative Filtering for Implicit Feedback Datasets", available at
 * [[http://dx.doi.org/10.1109/ICDM.2008.22]], adapted for the blocked approach used here.
 *
 * Essentially instead of finding the low-rank approximations to the rating matrix `R`,
 * this finds the approximations for a preference matrix `P` where the elements of `P` are 1 if
 * r > 0 and 0 if r <= 0. The ratings then act as 'confidence' values related to strength of
 * indicated user
 * preferences rather than explicit ratings given to items.
 */
@Since("0.8.0")
class ALS2 private (
    private var numUserBlocks: Int,
    private var numProductBlocks: Int,
    private var rank: Int,
    private var iterations: Int,
    private var lambda: Double,
    private var implicitPrefs: Boolean,
    private var alpha: Double,
    private var seed: Long = System.nanoTime()
  ) extends Serializable with Logging {

  /**
   * Constructs an ALS instance with default parameters: {numBlocks: -1, rank: 10, iterations: 10,
   * lambda: 0.01, implicitPrefs: false, alpha: 1.0}.
   */
  @Since("0.8.0")
  def this() = this(-1, -1, 10, 10, 0.01, false, 1.0)

  /** If true, do alternating nonnegative least squares. */
  private var nonnegative = false

  /** storage level for user/product in/out links */
  private var intermediateRDDStorageLevel: StorageLevel = StorageLevel.MEMORY_AND_DISK
  private var finalRDDStorageLevel: StorageLevel = StorageLevel.MEMORY_AND_DISK

  /** checkpoint interval */
  private var checkpointInterval: Int = 10

  /**
   * Set the number of blocks for both user blocks and product blocks to parallelize the computation
   * into; pass -1 for an auto-configured number of blocks. Default: -1.
   */
  @Since("0.8.0")
  def setBlocks(numBlocks: Int): this.type = {
    this.numUserBlocks = numBlocks
    this.numProductBlocks = numBlocks
    this
  }

  /**
   * Set the number of user blocks to parallelize the computation.
   */
  @Since("1.1.0")
  def setUserBlocks(numUserBlocks: Int): this.type = {
    this.numUserBlocks = numUserBlocks
    this
  }

  /**
   * Set the number of product blocks to parallelize the computation.
   */
  @Since("1.1.0")
  def setProductBlocks(numProductBlocks: Int): this.type = {
    this.numProductBlocks = numProductBlocks
    this
  }

  /** Set the rank of the feature matrices computed (number of features). Default: 10. */
  @Since("0.8.0")
  def setRank(rank: Int): this.type = {
    this.rank = rank
    this
  }

  /** Set the number of iterations to run. Default: 10. */
  @Since("0.8.0")
  def setIterations(iterations: Int): this.type = {
    this.iterations = iterations
    this
  }

  /** Set the regularization parameter, lambda. Default: 0.01. */
  @Since("0.8.0")
  def setLambda(lambda: Double): this.type = {
    this.lambda = lambda
    this
  }

  /** Sets whether to use implicit preference. Default: false. */
  @Since("0.8.1")
  def setImplicitPrefs(implicitPrefs: Boolean): this.type = {
    this.implicitPrefs = implicitPrefs
    this
  }

  /**
   * Sets the constant used in computing confidence in implicit ALS. Default: 1.0.
   */
  @Since("0.8.1")
  def setAlpha(alpha: Double): this.type = {
    this.alpha = alpha
    this
  }

  /** Sets a random seed to have deterministic results. */
  @Since("1.0.0")
  def setSeed(seed: Long): this.type = {
    this.seed = seed
    this
  }

  /**
   * Set whether the least-squares problems solved at each iteration should have
   * nonnegativity constraints.
   */
  @Since("1.1.0")
  def setNonnegative(b: Boolean): this.type = {
    this.nonnegative = b
    this
  }

  /**
   * :: DeveloperApi ::
   * Sets storage level for intermediate RDDs (user/product in/out links). The default value is
   * `MEMORY_AND_DISK`. Users can change it to a serialized storage, e.g., `MEMORY_AND_DISK_SER` and
   * set `spark.rdd.compress` to `true` to reduce the space requirement, at the cost of speed.
   */
  @DeveloperApi
  @Since("1.1.0")
  def setIntermediateRDDStorageLevel(storageLevel: StorageLevel): this.type = {
    require(storageLevel != StorageLevel.NONE,
      "ALS is not designed to run without persisting intermediate RDDs.")
    this.intermediateRDDStorageLevel = storageLevel
    this
  }

  /**
   * :: DeveloperApi ::
   * Sets storage level for final RDDs (user/product used in MatrixFactorizationModel). The default
   * value is `MEMORY_AND_DISK`. Users can change it to a serialized storage, e.g.
   * `MEMORY_AND_DISK_SER` and set `spark.rdd.compress` to `true` to reduce the space requirement,
   * at the cost of speed.
   */
  @DeveloperApi
  @Since("1.3.0")
  def setFinalRDDStorageLevel(storageLevel: StorageLevel): this.type = {
    this.finalRDDStorageLevel = storageLevel
    this
  }

  /**
   * Set period (in iterations) between checkpoints (default = 10). Checkpointing helps with
   * recovery (when nodes fail) and StackOverflow exceptions caused by long lineage. It also helps
   * with eliminating temporary shuffle files on disk, which can be important when there are many
   * ALS iterations. If the checkpoint directory is not set in [[org.apache.spark.SparkContext]],
   * this setting is ignored.
   */
  @DeveloperApi
  @Since("1.4.0")
  def setCheckpointInterval(checkpointInterval: Int): this.type = {
    this.checkpointInterval = checkpointInterval
    this
  }

  /**
   * Run ALS with the configured parameters on an input RDD of (user, product, rating) triples.
   * Returns a MatrixFactorizationModel with feature vectors for each user and product.
   */
  @Since("0.8.0")
  def run(ratings: RDD[Rating]): MatrixFactorizationModel = {
    val sc = ratings.context

    val numUserBlocks = if (this.numUserBlocks == -1) {
      math.max(sc.defaultParallelism, ratings.partitions.size / 2)
    } else {
      this.numUserBlocks
    }
    val numProductBlocks = if (this.numProductBlocks == -1) {
      math.max(sc.defaultParallelism, ratings.partitions.size / 2)
    } else {
      this.numProductBlocks
    }

    val (floatUserFactors, floatProdFactors) = NewALS2.train[Int](
      ratings = ratings.map(r => NewALS2.Rating(r.user, r.product, r.rating.toFloat)),
      rank = rank,
      numUserBlocks = numUserBlocks,
      numItemBlocks = numProductBlocks,
      maxIter = iterations,
      regParam = lambda,
      implicitPrefs = implicitPrefs,
      alpha = alpha,
      nonnegative = nonnegative,
      intermediateRDDStorageLevel = intermediateRDDStorageLevel,
      finalRDDStorageLevel = StorageLevel.NONE,
      checkpointInterval = checkpointInterval,
      seed = seed)

    val userFactors = floatUserFactors
      .mapValues(_.map(_.toDouble))
      .setName("users")
      .persist(finalRDDStorageLevel)
    val prodFactors = floatProdFactors
      .mapValues(_.map(_.toDouble))
      .setName("products")
      .persist(finalRDDStorageLevel)
    if (finalRDDStorageLevel != StorageLevel.NONE) {
      userFactors.count()
      prodFactors.count()
    }
    new MatrixFactorizationModel(rank, userFactors, prodFactors)
  }

  /**
   * Java-friendly version of [[ALS.run]].
   */
  @Since("1.3.0")
  def run(ratings: JavaRDD[Rating]): MatrixFactorizationModel = run(ratings.rdd)
}

/**
 * Top-level methods for calling Alternating Least Squares (ALS) matrix factorization.
 */
@Since("0.8.0")
object ALS2 {
  /**
   * Train a matrix factorization model given an RDD of ratings given by users to some products,
   * in the form of (userID, productID, rating) pairs. We approximate the ratings matrix as the
   * product of two lower-rank matrices of a given rank (number of features). To solve for these
   * features, we run a given number of iterations of ALS. This is done using a level of
   * parallelism given by `blocks`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   * @param blocks     level of parallelism to split computation into
   * @param seed       random seed
   */
  @Since("0.9.1")
  def train(
      ratings: RDD[Rating],
      rank: Int,
      iterations: Int,
      lambda: Double,
      blocks: Int,
      seed: Long
    ): MatrixFactorizationModel = {
    new ALS2(blocks, blocks, rank, iterations, lambda, false, 1.0, seed).run(ratings)
  }

  /**
   * Train a matrix factorization model given an RDD of ratings given by users to some products,
   * in the form of (userID, productID, rating) pairs. We approximate the ratings matrix as the
   * product of two lower-rank matrices of a given rank (number of features). To solve for these
   * features, we run a given number of iterations of ALS. This is done using a level of
   * parallelism given by `blocks`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   * @param blocks     level of parallelism to split computation into
   */
  @Since("0.8.0")
  def train(
      ratings: RDD[Rating],
      rank: Int,
      iterations: Int,
      lambda: Double,
      blocks: Int
    ): MatrixFactorizationModel = {
    new ALS2(blocks, blocks, rank, iterations, lambda, false, 1.0).run(ratings)
  }

  /**
   * Train a matrix factorization model given an RDD of ratings given by users to some products,
   * in the form of (userID, productID, rating) pairs. We approximate the ratings matrix as the
   * product of two lower-rank matrices of a given rank (number of features). To solve for these
   * features, we run a given number of iterations of ALS. The level of parallelism is determined
   * automatically based on the number of partitions in `ratings`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   */
  @Since("0.8.0")
  def train(ratings: RDD[Rating], rank: Int, iterations: Int, lambda: Double)
    : MatrixFactorizationModel = {
    train(ratings, rank, iterations, lambda, -1)
  }

  /**
   * Train a matrix factorization model given an RDD of ratings given by users to some products,
   * in the form of (userID, productID, rating) pairs. We approximate the ratings matrix as the
   * product of two lower-rank matrices of a given rank (number of features). To solve for these
   * features, we run a given number of iterations of ALS. The level of parallelism is determined
   * automatically based on the number of partitions in `ratings`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   */
  @Since("0.8.0")
  def train(ratings: RDD[Rating], rank: Int, iterations: Int)
    : MatrixFactorizationModel = {
    train(ratings, rank, iterations, 0.01, -1)
  }

  /**
   * Train a matrix factorization model given an RDD of 'implicit preferences' given by users
   * to some products, in the form of (userID, productID, preference) pairs. We approximate the
   * ratings matrix as the product of two lower-rank matrices of a given rank (number of features).
   * To solve for these features, we run a given number of iterations of ALS. This is done using
   * a level of parallelism given by `blocks`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   * @param blocks     level of parallelism to split computation into
   * @param alpha      confidence parameter
   * @param seed       random seed
   */
  @Since("0.8.1")
  def trainImplicit(
      ratings: RDD[Rating],
      rank: Int,
      iterations: Int,
      lambda: Double,
      blocks: Int,
      alpha: Double,
      seed: Long
    ): MatrixFactorizationModel = {
    new ALS2(blocks, blocks, rank, iterations, lambda, true, alpha, seed).run(ratings)
  }

  /**
   * Train a matrix factorization model given an RDD of 'implicit preferences' given by users
   * to some products, in the form of (userID, productID, preference) pairs. We approximate the
   * ratings matrix as the product of two lower-rank matrices of a given rank (number of features).
   * To solve for these features, we run a given number of iterations of ALS. This is done using
   * a level of parallelism given by `blocks`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   * @param blocks     level of parallelism to split computation into
   * @param alpha      confidence parameter
   */
  @Since("0.8.1")
  def trainImplicit(
      ratings: RDD[Rating],
      rank: Int,
      iterations: Int,
      lambda: Double,
      blocks: Int,
      alpha: Double
    ): MatrixFactorizationModel = {
    new ALS2(blocks, blocks, rank, iterations, lambda, true, alpha).run(ratings)
  }

  /**
   * Train a matrix factorization model given an RDD of 'implicit preferences' given by users to
   * some products, in the form of (userID, productID, preference) pairs. We approximate the
   * ratings matrix as the product of two lower-rank matrices of a given rank (number of features).
   * To solve for these features, we run a given number of iterations of ALS. The level of
   * parallelism is determined automatically based on the number of partitions in `ratings`.
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   * @param lambda     regularization factor (recommended: 0.01)
   * @param alpha      confidence parameter
   */
  @Since("0.8.1")
  def trainImplicit(ratings: RDD[Rating], rank: Int, iterations: Int, lambda: Double, alpha: Double)
    : MatrixFactorizationModel = {
    trainImplicit(ratings, rank, iterations, lambda, -1, alpha)
  }

  /**
   * Train a matrix factorization model given an RDD of 'implicit preferences' ratings given by
   * users to some products, in the form of (userID, productID, rating) pairs. We approximate the
   * ratings matrix as the product of two lower-rank matrices of a given rank (number of features).
   * To solve for these features, we run a given number of iterations of ALS. The level of
   * parallelism is determined automatically based on the number of partitions in `ratings`.
   * Model parameters `alpha` and `lambda` are set to reasonable default values
   *
   * @param ratings    RDD of (userID, productID, rating) pairs
   * @param rank       number of features to use
   * @param iterations number of iterations of ALS (recommended: 10-20)
   */
  @Since("0.8.1")
  def trainImplicit(ratings: RDD[Rating], rank: Int, iterations: Int)
    : MatrixFactorizationModel = {
    trainImplicit(ratings, rank, iterations, 0.01, -1, 1.0)
  }
}

ALSWithBias.scala

/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *    http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package org.apache.spark.ml.recommendation

import java.{ util => ju }
import java.io.IOException

import scala.collection.mutable
import scala.reflect.ClassTag
import scala.util.Sorting
import scala.util.hashing.byteswap64

import com.github.fommil.netlib.BLAS.{ getInstance => blas }
import org.apache.hadoop.fs.{ FileSystem, Path }
import org.json4s.DefaultFormats
import org.json4s.JsonDSL._

import org.apache.spark.{ Logging, Partitioner }
import org.apache.spark.annotation.{ Since, DeveloperApi, Experimental }
import org.apache.spark.ml.{ Estimator, Model }
import org.apache.spark.ml.param._
import org.apache.spark.ml.param.shared._
import org.apache.spark.ml.util._
import org.apache.spark.mllib.linalg.CholeskyDecomposition
import org.apache.spark.mllib.optimization.NNLS
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.DataFrame
import org.apache.spark.sql.functions._
import org.apache.spark.sql.types.{ DoubleType, FloatType, IntegerType, StructType }
import org.apache.spark.storage.StorageLevel
import org.apache.spark.util.Utils
import org.apache.spark.util.collection.{ OpenHashMap, OpenHashSet, SortDataFormat, Sorter }
import org.apache.spark.util.random.XORShiftRandom

/**
 * Common params for ALS and ALSModel2.
 */
private[recommendation] trait ALSModel2Params extends Params with HasPredictionCol {
  /**
   * Param for the column name for user ids.
   * Default: "user"
   * @group param
   */
  val userCol = new Param[String](this, "userCol", "column name for user ids")

  /** @group getParam */
  def getUserCol: String = $(userCol)

  /**
   * Param for the column name for item ids.
   * Default: "item"
   * @group param
   */
  val itemCol = new Param[String](this, "itemCol", "column name for item ids")

  /** @group getParam */
  def getItemCol: String = $(itemCol)
}

/**
 * Common params for ALS.
 */
private[recommendation] trait ALSParams extends ALSModel2Params with HasMaxIter with HasRegParam
    with HasPredictionCol with HasCheckpointInterval with HasSeed {

  /**
   * Param for rank of the matrix factorization (>= 1).
   * Default: 10
   * @group param
   */
  val rank = new IntParam(this, "rank", "rank of the factorization", ParamValidators.gtEq(1))

  /** @group getParam */
  def getRank: Int = $(rank)

  /**
   * Param for number of user blocks (>= 1).
   * Default: 10
   * @group param
   */
  val numUserBlocks = new IntParam(this, "numUserBlocks", "number of user blocks",
    ParamValidators.gtEq(1))

  /** @group getParam */
  def getNumUserBlocks: Int = $(numUserBlocks)

  /**
   * Param for number of item blocks (>= 1).
   * Default: 10
   * @group param
   */
  val numItemBlocks = new IntParam(this, "numItemBlocks", "number of item blocks",
    ParamValidators.gtEq(1))

  /** @group getParam */
  def getNumItemBlocks: Int = $(numItemBlocks)

  /**
   * Param to decide whether to use implicit preference.
   * Default: false
   * @group param
   */
  val implicitPrefs = new BooleanParam(this, "implicitPrefs", "whether to use implicit preference")

  /** @group getParam */
  def getImplicitPrefs: Boolean = $(implicitPrefs)

  /**
   * Param for the alpha parameter in the implicit preference formulation (>= 0).
   * Default: 1.0
   * @group param
   */
  val alpha = new DoubleParam(this, "alpha", "alpha for implicit preference",
    ParamValidators.gtEq(0))

  /** @group getParam */
  def getAlpha: Double = $(alpha)

  /**
   * Param for the column name for ratings.
   * Default: "rating"
   * @group param
   */
  val ratingCol = new Param[String](this, "ratingCol", "column name for ratings")

  /** @group getParam */
  def getRatingCol: String = $(ratingCol)

  /**
   * Param for whether to apply nonnegativity constraints.
   * Default: false
   * @group param
   */
  val nonnegative = new BooleanParam(
    this, "nonnegative", "whether to use nonnegative constraint for least squares")

  /** @group getParam */
  def getNonnegative: Boolean = $(nonnegative)

  setDefault(rank -> 10, maxIter -> 10, regParam -> 0.1, numUserBlocks -> 10, numItemBlocks -> 10,
    implicitPrefs -> false, alpha -> 1.0, userCol -> "user", itemCol -> "item",
    ratingCol -> "rating", nonnegative -> false, checkpointInterval -> 10)

  /**
   * Validates and transforms the input schema.
   * @param schema input schema
   * @return output schema
   */
  protected def validateAndTransformSchema(schema: StructType): StructType = {
    SchemaUtils.checkColumnType(schema, $(userCol), IntegerType)
    SchemaUtils.checkColumnType(schema, $(itemCol), IntegerType)
    val ratingType = schema($(ratingCol)).dataType
    require(ratingType == FloatType || ratingType == DoubleType)
    SchemaUtils.appendColumn(schema, $(predictionCol), FloatType)
  }
}

/**
 * :: Experimental ::
 * Model fitted by ALS.
 *
 * @param rank rank of the matrix factorization model
 * @param userFactors a DataFrame that stores user factors in two columns: `id` and `features`
 * @param itemFactors a DataFrame that stores item factors in two columns: `id` and `features`
 */
@Experimental
class ALSModel2 private[ml] (
  override val uid: String,
  val rank: Int,
  @transient val userFactors: DataFrame,
  @transient val itemFactors: DataFrame)
    extends Model[ALSModel2] with ALSModel2Params with MLWritable {

  /** @group setParam */
  def setUserCol(value: String): this.type = set(userCol, value)

  /** @group setParam */
  def setItemCol(value: String): this.type = set(itemCol, value)

  /** @group setParam */
  def setPredictionCol(value: String): this.type = set(predictionCol, value)

  override def transform(dataset: DataFrame): DataFrame = {
    // Register a UDF for DataFrame, and then
    // create a new column named map(predictionCol) by running the predict UDF.
    val predict = udf { (userFeatures: Seq[Float], itemFeatures: Seq[Float]) =>
      if (userFeatures != null && itemFeatures != null) {
        blas.sdot(rank, userFeatures.toArray, 1, itemFeatures.toArray, 1)
      } else {
        Float.NaN
      }
    }
    dataset
      .join(userFactors, dataset($(userCol)) === userFactors("id"), "left")
      .join(itemFactors, dataset($(itemCol)) === itemFactors("id"), "left")
      .select(dataset("*"),
        predict(userFactors("features"), itemFactors("features")).as($(predictionCol)))
  }

  override def transformSchema(schema: StructType): StructType = {
    SchemaUtils.checkColumnType(schema, $(userCol), IntegerType)
    SchemaUtils.checkColumnType(schema, $(itemCol), IntegerType)
    SchemaUtils.appendColumn(schema, $(predictionCol), FloatType)
  }

  override def copy(extra: ParamMap): ALSModel2 = {
    val copied = new ALSModel2(uid, rank, userFactors, itemFactors)
    copyValues(copied, extra).setParent(parent)
  }

  @Since("1.6.0")
  override def write: MLWriter = new ALSModel2.ALSModel2Writer(this)
}

@Since("1.6.0")
object ALSModel2 extends MLReadable[ALSModel2] {

  @Since("1.6.0")
  override def read: MLReader[ALSModel2] = new ALSModel2Reader

  @Since("1.6.0")
  override def load(path: String): ALSModel2 = super.load(path)

  private[ALSModel2] class ALSModel2Writer(instance: ALSModel2) extends MLWriter {

    override protected def saveImpl(path: String): Unit = {
      val extraMetadata = "rank" -> instance.rank
      DefaultParamsWriter.saveMetadata(instance, path, sc, Some(extraMetadata))
      val userPath = new Path(path, "userFactors").toString
      instance.userFactors.write.format("parquet").save(userPath)
      val itemPath = new Path(path, "itemFactors").toString
      instance.itemFactors.write.format("parquet").save(itemPath)
    }
  }

  private class ALSModel2Reader extends MLReader[ALSModel2] {

    /** Checked against metadata when loading model */
    private val className = classOf[ALSModel2].getName

    override def load(path: String): ALSModel2 = {
      val metadata = DefaultParamsReader.loadMetadata(path, sc, className)
      implicit val format = DefaultFormats
      val rank = (metadata.metadata \ "rank").extract[Int]
      val userPath = new Path(path, "userFactors").toString
      val userFactors = sqlContext.read.format("parquet").load(userPath)
      val itemPath = new Path(path, "itemFactors").toString
      val itemFactors = sqlContext.read.format("parquet").load(itemPath)

      val model = new ALSModel2(metadata.uid, rank, userFactors, itemFactors)

      DefaultParamsReader.getAndSetParams(model, metadata)
      model
    }
  }
}

/**
 * :: Experimental ::
 * Alternating Least Squares (ALS) matrix factorization.
 *
 * ALS attempts to estimate the ratings matrix `R` as the product of two lower-rank matrices,
 * `X` and `Y`, i.e. `X * Yt = R`. Typically these approximations are called 'factor' matrices.
 * The general approach is iterative. During each iteration, one of the factor matrices is held
 * constant, while the other is solved for using least squares. The newly-solved factor matrix is
 * then held constant while solving for the other factor matrix.
 *
 * This is a blocked implementation of the ALS factorization algorithm that groups the two sets
 * of factors (referred to as "users" and "products") into blocks and reduces communication by only
 * sending one copy of each user vector to each product block on each iteration, and only for the
 * product blocks that need that user's feature vector. This is achieved by pre-computing some
 * information about the ratings matrix to determine the "out-links" of each user (which blocks of
 * products it will contribute to) and "in-link" information for each product (which of the feature
 * vectors it receives from each user block it will depend on). This allows us to send only an
 * array of feature vectors between each user block and product block, and have the product block
 * find the users' ratings and update the products based on these messages.
 *
 * For implicit preference data, the algorithm used is based on
 * "Collaborative Filtering for Implicit Feedback Datasets", available at
 * [[http://dx.doi.org/10.1109/ICDM.2008.22]], adapted for the blocked approach used here.
 *
 * Essentially instead of finding the low-rank approximations to the rating matrix `R`,
 * this finds the approximations for a preference matrix `P` where the elements of `P` are 1 if
 * r > 0 and 0 if r <= 0. The ratings then act as 'confidence' values related to strength of
 * indicated user
 * preferences rather than explicit ratings given to items.
 */
@Experimental
class ALS2(override val uid: String) extends Estimator[ALSModel2] with ALSParams
    with DefaultParamsWritable {

  import org.apache.spark.ml.recommendation.ALS2.Rating

  def this() = this(Identifiable.randomUID("als"))

  /** @group setParam */
  def setRank(value: Int): this.type = set(rank, value)

  /** @group setParam */
  def setNumUserBlocks(value: Int): this.type = set(numUserBlocks, value)

  /** @group setParam */
  def setNumItemBlocks(value: Int): this.type = set(numItemBlocks, value)

  /** @group setParam */
  def setImplicitPrefs(value: Boolean): this.type = set(implicitPrefs, value)

  /** @group setParam */
  def setAlpha(value: Double): this.type = set(alpha, value)

  /** @group setParam */
  def setUserCol(value: String): this.type = set(userCol, value)

  /** @group setParam */
  def setItemCol(value: String): this.type = set(itemCol, value)

  /** @group setParam */
  def setRatingCol(value: String): this.type = set(ratingCol, value)

  /** @group setParam */
  def setPredictionCol(value: String): this.type = set(predictionCol, value)

  /** @group setParam */
  def setMaxIter(value: Int): this.type = set(maxIter, value)

  /** @group setParam */
  def setRegParam(value: Double): this.type = set(regParam, value)

  /** @group setParam */
  def setNonnegative(value: Boolean): this.type = set(nonnegative, value)

  /** @group setParam */
  def setCheckpointInterval(value: Int): this.type = set(checkpointInterval, value)

  /** @group setParam */
  def setSeed(value: Long): this.type = set(seed, value)

  /**
   * Sets both numUserBlocks and numItemBlocks to the specific value.
   * @group setParam
   */
  def setNumBlocks(value: Int): this.type = {
    setNumUserBlocks(value)
    setNumItemBlocks(value)
    this
  }

  override def fit(dataset: DataFrame): ALSModel2 = {
    import dataset.sqlContext.implicits._
    val r = if ($(ratingCol) != "") col($(ratingCol)).cast(FloatType) else lit(1.0f)
    val ratings = dataset
      .select(col($(userCol)).cast(IntegerType), col($(itemCol)).cast(IntegerType), r)
      .map { row =>
        Rating(row.getInt(0), row.getInt(1), row.getFloat(2))
      }
    val (userFactors, itemFactors) = ALS2.train(ratings, rank = $(rank),
      numUserBlocks = $(numUserBlocks), numItemBlocks = $(numItemBlocks),
      maxIter = $(maxIter), regParam = $(regParam), implicitPrefs = $(implicitPrefs),
      alpha = $(alpha), nonnegative = $(nonnegative),
      checkpointInterval = $(checkpointInterval), seed = $(seed))
    val userDF = userFactors.toDF("id", "features")
    val itemDF = itemFactors.toDF("id", "features")
    val model = new ALSModel2(uid, $(rank), userDF, itemDF).setParent(this)
    copyValues(model)
  }

  override def transformSchema(schema: StructType): StructType = {
    validateAndTransformSchema(schema)
  }

  override def copy(extra: ParamMap): ALS2 = defaultCopy(extra)
}

/**
 * :: DeveloperApi ::
 * An implementation of ALS that supports generic ID types, specialized for Int and Long. This is
 * exposed as a developer API for users who do need other ID types. But it is not recommended
 * because it increases the shuffle size and memory requirement during training. For simplicity,
 * users and items must have the same type. The number of distinct users/items should be smaller
 * than 2 billion.
 */
@DeveloperApi
object ALS2 extends DefaultParamsReadable[ALS2] with Logging {

  /**
   * :: DeveloperApi ::
   * Rating class for better code readability.
   */
  @DeveloperApi
  case class Rating[@specialized(Int, Long) ID](user: ID, item: ID, rating: Float)

  //  @Since("1.6.0")
  //  override def load(path: String): ALS2 = super.load(path)

  /** Trait for least squares solvers applied to the normal equation. */
  private[recommendation] trait LeastSquaresNESolver extends Serializable {
    /** Solves a least squares problem with regularization (possibly with other constraints). */
    def solve(ne: NormalEquation, lambda: Double): Array[Float]
  }

  /** Cholesky solver for least square problems. */
  private[recommendation] class CholeskySolver extends LeastSquaresNESolver {

    /**
     * Solves a least squares problem with L2 regularization:
     *
     *   min norm(A x - b)^2^ + lambda * norm(x)^2^
     *
     * @param ne a [[NormalEquation]] instance that contains AtA, Atb, and n (number of instances)
     * @param lambda regularization constant
     * @return the solution x
     */
    override def solve(ne: NormalEquation, lambda: Double): Array[Float] = {
      val k = ne.k
      // Add scaled lambda to the diagonals of AtA.
      var i = 0
      var j = 2
      while (i < ne.triK) {
        ne.ata(i) += lambda
        i += j
        j += 1
      }
      CholeskyDecomposition.solve(ne.ata, ne.atb)
      val x = new Array[Float](k)
      i = 0
      while (i < k) {
        x(i) = ne.atb(i).toFloat
        i += 1
      }
      ne.reset()
      x
    }
  }

  /** NNLS solver. */
  private[recommendation] class NNLSSolver extends LeastSquaresNESolver {
    private var rank: Int = -1
    private var workspace: NNLS.Workspace = _
    private var ata: Array[Double] = _
    private var initialized: Boolean = false

    private def initialize(rank: Int): Unit = {
      if (!initialized) {
        this.rank = rank
        workspace = NNLS.createWorkspace(rank)
        ata = new Array[Double](rank * rank)
        initialized = true
      } else {
        require(this.rank == rank)
      }
    }

    /**
     * Solves a nonnegative least squares problem with L2 regularizatin:
     *
     *   min_x_  norm(A x - b)^2^ + lambda * n * norm(x)^2^
     *   subject to x >= 0
     */
    override def solve(ne: NormalEquation, lambda: Double): Array[Float] = {
      val rank = ne.k
      initialize(rank)
      fillAtA(ne.ata, lambda)
      val x = NNLS.solve(ata, ne.atb, workspace)
      ne.reset()
      x.map(x => x.toFloat)
    }

    /**
     * Given a triangular matrix in the order of fillXtX above, compute the full symmetric square
     * matrix that it represents, storing it into destMatrix.
     */
    private def fillAtA(triAtA: Array[Double], lambda: Double) {
      var i = 0
      var pos = 0
      var a = 0.0
      while (i < rank) {
        var j = 0
        while (j <= i) {
          a = triAtA(pos)
          ata(i * rank + j) = a
          ata(j * rank + i) = a
          pos += 1
          j += 1
        }
        ata(i * rank + i) += lambda
        i += 1
      }
    }
  }

  /**
   * Representing a normal equation to solve the following weighted least squares problem:
   *
   * minimize \sum,,i,, c,,i,, (a,,i,,^T^ x - b,,i,,)^2^ + lambda * x^T^ x.
   *
   * Its normal equation is given by
   *
   * \sum,,i,, c,,i,, (a,,i,, a,,i,,^T^ x - b,,i,, a,,i,,) + lambda * x = 0.
   */
  private[recommendation] class NormalEquation(val k: Int) extends Serializable {

    /** Number of entries in the upper triangular part of a k-by-k matrix. */
    val triK = k * (k + 1) / 2
    /** A^T^ * A */
    val ata = new Array[Double](triK)
    /** A^T^ * b */
    val atb = new Array[Double](k)

    private val da = new Array[Double](k)
    private val upper = "U"

    private def copyToDouble(a: Array[Float]): Unit = {
      var i = 0
      while (i < k) {
        da(i) = a(i)
        i += 1
      }
    }

    /** Adds an observation. */
    def add(a: Array[Float], b: Double, c: Double = 1.0): this.type = {
      require(c >= 0.0)
      require(a.length == k, "Check: a = " + a.length + ", k = " + k)
      copyToDouble(a)
      blas.dspr(upper, k, c, da, 1, ata)
      if (b != 0.0) {
        blas.daxpy(k, c * b, da, 1, atb, 1)
      }
      this
    }

    /** Merges another normal equation object. */
    def merge(other: NormalEquation): this.type = {
      require(other.k == k)
      blas.daxpy(ata.length, 1.0, other.ata, 1, ata, 1)
      blas.daxpy(atb.length, 1.0, other.atb, 1, atb, 1)
      this
    }

    /** Resets everything to zero, which should be called after each solve. */
    def reset(): Unit = {
      ju.Arrays.fill(ata, 0.0)
      ju.Arrays.fill(atb, 0.0)
    }
  }

  /**
   * :: DeveloperApi ::
   * Implementation of the ALS algorithm.
   */
  @DeveloperApi
  def train[ID: ClassTag]( // scalastyle:ignore
    ratings: RDD[Rating[ID]],
    rank: Int = 10,
    numUserBlocks: Int = 10,
    numItemBlocks: Int = 10,
    maxIter: Int = 10,
    regParam: Double = 1.0,
    implicitPrefs: Boolean = false,
    alpha: Double = 1.0,
    nonnegative: Boolean = false,
    intermediateRDDStorageLevel: StorageLevel = StorageLevel.MEMORY_AND_DISK,
    finalRDDStorageLevel: StorageLevel = StorageLevel.MEMORY_AND_DISK,
    checkpointInterval: Int = 10,
    seed: Long = 0L)(
      implicit ord: Ordering[ID]): (RDD[(ID, Array[Float])], RDD[(ID, Array[Float])]) = {
    require(intermediateRDDStorageLevel != StorageLevel.NONE,
      "ALS is not designed to run without persisting intermediate RDDs.")
    val sc = ratings.sparkContext
    val userPart = new ALSPartitioner(numUserBlocks)
    val itemPart = new ALSPartitioner(numItemBlocks)
    val userLocalIndexEncoder = new LocalIndexEncoder(userPart.numPartitions)
    val itemLocalIndexEncoder = new LocalIndexEncoder(itemPart.numPartitions)
    val solver = if (nonnegative) new NNLSSolver else new CholeskySolver
    val blockRatings = partitionRatings(ratings, userPart, itemPart)
      .persist(intermediateRDDStorageLevel)
    val (userInBlocks, userOutBlocks) =
      makeBlocks("user", blockRatings, userPart, itemPart, intermediateRDDStorageLevel)
    // materialize blockRatings and user blocks
    userOutBlocks.count()
    val swappedBlockRatings = blockRatings.map {
      case ((userBlockId, itemBlockId), RatingBlock(userIds, itemIds, localRatings)) =>
        ((itemBlockId, userBlockId), RatingBlock(itemIds, userIds, localRatings))
    }
    val (itemInBlocks, itemOutBlocks) =
      makeBlocks("item", swappedBlockRatings, itemPart, userPart, intermediateRDDStorageLevel)
    // materialize item blocks
    itemOutBlocks.count()
    val seedGen = new XORShiftRandom(seed)

    //add bias
    var userFactors = initialize(userInBlocks, rank + 1, seedGen.nextLong())
    var itemFactors = initialize(itemInBlocks, rank + 1, seedGen.nextLong())

    var previousCheckpointFile: Option[String] = None
    val shouldCheckpoint: Int => Boolean = (iter) =>
      sc.checkpointDir.isDefined && checkpointInterval != -1 && (iter % checkpointInterval == 0)
    val deletePreviousCheckpointFile: () => Unit = () =>
      previousCheckpointFile.foreach { file =>
        try {
          FileSystem.get(sc.hadoopConfiguration).delete(new Path(file), true)
        } catch {
          case e: IOException =>
            logWarning(s"Cannot delete checkpoint file $file:", e)
        }
      }
    if (implicitPrefs) {
      for (iter <- 1 to maxIter) {
        userFactors.setName(s"userFactors-$iter").persist(intermediateRDDStorageLevel)
        val previousItemFactors = itemFactors
        var biases = itemFactors.map {
          case (blockId, factors) =>
            val b = factors.map { x => x(0) }
            (blockId, b)
        }
        itemFactors = computeFactors(biases, userFactors, userOutBlocks, itemInBlocks, rank, regParam,
          userLocalIndexEncoder, implicitPrefs, alpha, solver)
        previousItemFactors.unpersist()
        itemFactors.setName(s"itemFactors-$iter").persist(intermediateRDDStorageLevel)
        // TODO: Generalize PeriodicGraphCheckpointer and use it here.
        if (shouldCheckpoint(iter)) {
          itemFactors.checkpoint() // itemFactors gets materialized in computeFactors.
        }
        val previousUserFactors = userFactors

        biases = userFactors.map {
          case (blockId, factors) =>
            val b = factors.map(x => x(0))
            (blockId, b)
        }
        userFactors = computeFactors(biases, itemFactors, itemOutBlocks, userInBlocks, rank, regParam,
          itemLocalIndexEncoder, implicitPrefs, alpha, solver)
        if (shouldCheckpoint(iter)) {
          deletePreviousCheckpointFile()
          previousCheckpointFile = itemFactors.getCheckpointFile
        }
        previousUserFactors.unpersist()
      }
    } else {
      for (iter <- 0 until maxIter) {
        var biases = itemFactors.map {
          case (blockId, factors) =>
            val b = factors.map { x => x(0) }
            (blockId, b)
        }
        itemFactors = computeFactors(biases, userFactors, userOutBlocks, itemInBlocks, rank, regParam,
          userLocalIndexEncoder, solver = solver)
        if (shouldCheckpoint(iter)) {
          itemFactors.checkpoint()
          itemFactors.count() // checkpoint item factors and cut lineage
          deletePreviousCheckpointFile()
          previousCheckpointFile = itemFactors.getCheckpointFile
        }

        biases = userFactors.map {
          case (blockId, factors) =>
            val b = factors.map { x => x(0) }
            (blockId, b)
        }
        userFactors = computeFactors(biases, itemFactors, itemOutBlocks, userInBlocks, rank, regParam,
          itemLocalIndexEncoder, solver = solver)
      }
    }
    val userIdAndFactors = userInBlocks
      .mapValues(_.srcIds)
      .join(userFactors)
      .mapPartitions({ items =>
        items.flatMap {
          case (_, (ids, factors)) =>
            ids.view.zip(factors)
        }
        // Preserve the partitioning because IDs are consistent with the partitioners in userInBlocks
        // and userFactors.
      }, preservesPartitioning = true)
      .map(v => (v._1, v._2.slice(1, v._2.length)))
      .setName("userFactors")
      .persist(finalRDDStorageLevel)
    val itemIdAndFactorsWithBiases = itemInBlocks
      .mapValues(_.srcIds)
      .join(itemFactors)
      .mapPartitions({ items =>
        items.flatMap {
          case (_, (ids, factors)) =>
            ids.view.zip(factors)
        }
      }, preservesPartitioning = true)
      .map(v => (v._1, (v._2(0), v._2.slice(1, v._2.length))))
      .setName("itemFactors")
      .persist(finalRDDStorageLevel)

    val biasSamples1 = itemIdAndFactorsWithBiases.map(x => (x._1, x._2._1))
      .filter(_._1.asInstanceOf[Int] > 0).take(50)
    val biasSamples2 = itemIdAndFactorsWithBiases.map(x => (x._1, x._2._1))
      .filter(_._1.asInstanceOf[Int] < 0).take(50)
    println(biasSamples1.mkString("\n------------------------------------\n"))
    println("=============================================================")
    println(biasSamples2.mkString("\n------------------------------------\n"))
    val itemIdAndFactors = itemIdAndFactorsWithBiases.map(x => (x._1, x._2._2))

    if (finalRDDStorageLevel != StorageLevel.NONE) {
      userIdAndFactors.count()
      itemFactors.unpersist()
      itemIdAndFactors.count()
      userInBlocks.unpersist()
      userOutBlocks.unpersist()
      itemInBlocks.unpersist()
      itemOutBlocks.unpersist()
      blockRatings.unpersist()
    }
    (userIdAndFactors, itemIdAndFactors)
  }

  /**
   * Factor block that stores factors (Array[Float]) in an Array.
   */
  private type FactorBlock = Array[Array[Float]]
  private type FactorBlock2 = Array[(Array[Float], Float)]

  /**
   * Out-link block that stores, for each dst (item/user) block, which src (user/item) factors to
   * send. For example, outLinkBlock(0) contains the local indices (not the original src IDs) of the
   * src factors in this block to send to dst block 0.
   */
  private type OutBlock = Array[Array[Int]]

  /**
   * In-link block for computing src (user/item) factors. This includes the original src IDs
   * of the elements within this block as well as encoded dst (item/user) indices and corresponding
   * ratings. The dst indices are in the form of (blockId, localIndex), which are not the original
   * dst IDs. To compute src factors, we expect receiving dst factors that match the dst indices.
   * For example, if we have an in-link record
   *
   * {srcId: 0, dstBlockId: 2, dstLocalIndex: 3, rating: 5.0},
   *
   * and assume that the dst factors are stored as dstFactors: Map[Int, Array[Array[Float]]], which
   * is a blockId to dst factors map, the corresponding dst factor of the record is dstFactor(2)(3).
   *
   * We use a CSC-like (compressed sparse column) format to store the in-link information. So we can
   * compute src factors one after another using only one normal equation instance.
   *
   * @param srcIds src ids (ordered)
   * @param dstPtrs dst pointers. Elements in range [dstPtrs(i), dstPtrs(i+1)) of dst indices and
   *                ratings are associated with srcIds(i).
   * @param dstEncodedIndices encoded dst indices
   * @param ratings ratings
   *
   * @see [[LocalIndexEncoder]]
   */
  private[recommendation] case class InBlock[@specialized(Int, Long) ID: ClassTag](
      srcIds: Array[ID],
      dstPtrs: Array[Int],
      dstEncodedIndices: Array[Int],
      ratings: Array[Float]) {
    /** Size of the block. */
    def size: Int = ratings.length
    require(dstEncodedIndices.length == size)
    require(dstPtrs.length == srcIds.length + 1)
  }

  /**
   * Initializes factors randomly given the in-link blocks.
   *
   * @param inBlocks in-link blocks
   * @param rank rank
   * @return initialized factor blocks
   */
  private def initialize[ID](
    inBlocks: RDD[(Int, InBlock[ID])],
    rank: Int,
    seed: Long): RDD[(Int, FactorBlock)] = {
    // Choose a unit vector uniformly at random from the unit sphere, but from the
    // "first quadrant" where all elements are nonnegative. This can be done by choosing
    // elements distributed as Normal(0,1) and taking the absolute value, and then normalizing.
    // This appears to create factorizations that have a slightly better reconstruction
    // (<1%) compared picking elements uniformly at random in [0,1].
    inBlocks.map {
      case (srcBlockId, inBlock) =>
        val random = new XORShiftRandom(byteswap64(seed ^ srcBlockId))
        val factors = Array.fill(inBlock.srcIds.length) {
          val factor = Array.fill(rank)(random.nextGaussian().toFloat)
          val nrm = blas.snrm2(rank, factor, 1)
          blas.sscal(rank, 1.0f / nrm, factor, 1)
          factor(0) = 0.0f //bias = 0
          factor
        }
        (srcBlockId, factors)
    }
  }

  /**
   * A rating block that contains src IDs, dst IDs, and ratings, stored in primitive arrays.
   */
  private[recommendation] case class RatingBlock[@specialized(Int, Long) ID: ClassTag](
      srcIds: Array[ID],
      dstIds: Array[ID],
      ratings: Array[Float]) {
    /** Size of the block. */
    def size: Int = srcIds.length
    require(dstIds.length == srcIds.length)
    require(ratings.length == srcIds.length)
  }

  /**
   * Builder for [[RatingBlock]]. [[mutable.ArrayBuilder]] is used to avoid boxing/unboxing.
   */
  private[recommendation] class RatingBlockBuilder[@specialized(Int, Long) ID: ClassTag]
      extends Serializable {

    private val srcIds = mutable.ArrayBuilder.make[ID]
    private val dstIds = mutable.ArrayBuilder.make[ID]
    private val ratings = mutable.ArrayBuilder.make[Float]
    var size = 0

    /** Adds a rating. */
    def add(r: Rating[ID]): this.type = {
      size += 1
      srcIds += r.user
      dstIds += r.item
      ratings += r.rating
      this
    }

    /** Merges another [[RatingBlockBuilder]]. */
    def merge(other: RatingBlock[ID]): this.type = {
      size += other.srcIds.length
      srcIds ++= other.srcIds
      dstIds ++= other.dstIds
      ratings ++= other.ratings
      this
    }

    /** Builds a [[RatingBlock]]. */
    def build(): RatingBlock[ID] = {
      RatingBlock[ID](srcIds.result(), dstIds.result(), ratings.result())
    }
  }

  /**
   * Partitions raw ratings into blocks.
   *
   * @param ratings raw ratings
   * @param srcPart partitioner for src IDs
   * @param dstPart partitioner for dst IDs
   *
   * @return an RDD of rating blocks in the form of ((srcBlockId, dstBlockId), ratingBlock)
   */
  private def partitionRatings[ID: ClassTag](
    ratings: RDD[Rating[ID]],
    srcPart: Partitioner,
    dstPart: Partitioner): RDD[((Int, Int), RatingBlock[ID])] = {

    /* The implementation produces the same result as the following but generates less objects.

     ratings.map { r =>
       ((srcPart.getPartition(r.user), dstPart.getPartition(r.item)), r)
     }.aggregateByKey(new RatingBlockBuilder)(
         seqOp = (b, r) => b.add(r),
         combOp = (b0, b1) => b0.merge(b1.build()))
       .mapValues(_.build())
     */

    val numPartitions = srcPart.numPartitions * dstPart.numPartitions
    ratings.mapPartitions { iter =>
      val builders = Array.fill(numPartitions)(new RatingBlockBuilder[ID])
      iter.flatMap { r =>
        val srcBlockId = srcPart.getPartition(r.user)
        val dstBlockId = dstPart.getPartition(r.item)
        val idx = srcBlockId + srcPart.numPartitions * dstBlockId
        val builder = builders(idx)
        builder.add(r)
        if (builder.size >= 2048) { // 2048 * (3 * 4) = 24k
          builders(idx) = new RatingBlockBuilder
          Iterator.single(((srcBlockId, dstBlockId), builder.build()))
        } else {
          Iterator.empty
        }
      } ++ {
        builders.view.zipWithIndex.filter(_._1.size > 0).map {
          case (block, idx) =>
            val srcBlockId = idx % srcPart.numPartitions
            val dstBlockId = idx / srcPart.numPartitions
            ((srcBlockId, dstBlockId), block.build())
        }
      }
    }.groupByKey().mapValues { blocks =>
      val builder = new RatingBlockBuilder[ID]
      blocks.foreach(builder.merge)
      builder.build()
    }.setName("ratingBlocks")
  }

  /**
   * Builder for uncompressed in-blocks of (srcId, dstEncodedIndex, rating) tuples.
   * @param encoder encoder for dst indices
   */
  private[recommendation] class UncompressedInBlockBuilder[@specialized(Int, Long) ID: ClassTag](
      encoder: LocalIndexEncoder)(
          implicit ord: Ordering[ID]) {

    private val srcIds = mutable.ArrayBuilder.make[ID]
    private val dstEncodedIndices = mutable.ArrayBuilder.make[Int]
    private val ratings = mutable.ArrayBuilder.make[Float]

    /**
     * Adds a dst block of (srcId, dstLocalIndex, rating) tuples.
     *
     * @param dstBlockId dst block ID
     * @param srcIds original src IDs
     * @param dstLocalIndices dst local indices
     * @param ratings ratings
     */
    def add(
      dstBlockId: Int,
      srcIds: Array[ID],
      dstLocalIndices: Array[Int],
      ratings: Array[Float]): this.type = {
      val sz = srcIds.length
      require(dstLocalIndices.length == sz)
      require(ratings.length == sz)
      this.srcIds ++= srcIds
      this.ratings ++= ratings
      var j = 0
      while (j < sz) {
        this.dstEncodedIndices += encoder.encode(dstBlockId, dstLocalIndices(j))
        j += 1
      }
      this
    }

    /** Builds a [[UncompressedInBlock]]. */
    def build(): UncompressedInBlock[ID] = {
      new UncompressedInBlock(srcIds.result(), dstEncodedIndices.result(), ratings.result())
    }
  }

  /**
   * A block of (srcId, dstEncodedIndex, rating) tuples stored in primitive arrays.
   */
  private[recommendation] class UncompressedInBlock[@specialized(Int, Long) ID: ClassTag](
      val srcIds: Array[ID],
      val dstEncodedIndices: Array[Int],
      val ratings: Array[Float])(
          implicit ord: Ordering[ID]) {

    /** Size the of block. */
    def length: Int = srcIds.length

    /**
     * Compresses the block into an [[InBlock]]. The algorithm is the same as converting a
     * sparse matrix from coordinate list (COO) format into compressed sparse column (CSC) format.
     * Sorting is done using Spark's built-in Timsort to avoid generating too many objects.
     */
    def compress(): InBlock[ID] = {
      val sz = length
      assert(sz > 0, "Empty in-link block should not exist.")
      sort()
      val uniqueSrcIdsBuilder = mutable.ArrayBuilder.make[ID]
      val dstCountsBuilder = mutable.ArrayBuilder.make[Int]
      var preSrcId = srcIds(0)
      uniqueSrcIdsBuilder += preSrcId
      var curCount = 1
      var i = 1
      var j = 0
      while (i < sz) {
        val srcId = srcIds(i)
        if (srcId != preSrcId) {
          uniqueSrcIdsBuilder += srcId
          dstCountsBuilder += curCount
          preSrcId = srcId
          j += 1
          curCount = 0
        }
        curCount += 1
        i += 1
      }
      dstCountsBuilder += curCount
      val uniqueSrcIds = uniqueSrcIdsBuilder.result()
      val numUniqueSrdIds = uniqueSrcIds.length
      val dstCounts = dstCountsBuilder.result()
      val dstPtrs = new Array[Int](numUniqueSrdIds + 1)
      var sum = 0
      i = 0
      while (i < numUniqueSrdIds) {
        sum += dstCounts(i)
        i += 1
        dstPtrs(i) = sum
      }
      InBlock(uniqueSrcIds, dstPtrs, dstEncodedIndices, ratings)
    }

    private def sort(): Unit = {
      val sz = length
      // Since there might be interleaved log messages, we insert a unique id for easy pairing.
      val sortId = Utils.random.nextInt()
      logDebug(s"Start sorting an uncompressed in-block of size $sz. (sortId = $sortId)")
      val start = System.nanoTime()
      val sorter = new Sorter(new UncompressedInBlockSort[ID])
      sorter.sort(this, 0, length, Ordering[KeyWrapper[ID]])
      val duration = (System.nanoTime() - start) / 1e9
      logDebug(s"Sorting took $duration seconds. (sortId = $sortId)")
    }
  }

  /**
   * A wrapper that holds a primitive key.
   *
   * @see [[UncompressedInBlockSort]]
   */
  private class KeyWrapper[@specialized(Int, Long) ID: ClassTag](
      implicit ord: Ordering[ID]) extends Ordered[KeyWrapper[ID]] {

    var key: ID = _

    override def compare(that: KeyWrapper[ID]): Int = {
      ord.compare(key, that.key)
    }

    def setKey(key: ID): this.type = {
      this.key = key
      this
    }
  }

  /**
   * [[SortDataFormat]] of [[UncompressedInBlock]] used by [[Sorter]].
   */
  private class UncompressedInBlockSort[@specialized(Int, Long) ID: ClassTag](
    implicit ord: Ordering[ID])
      extends SortDataFormat[KeyWrapper[ID], UncompressedInBlock[ID]] {

    override def newKey(): KeyWrapper[ID] = new KeyWrapper()

    override def getKey(
      data: UncompressedInBlock[ID],
      pos: Int,
      reuse: KeyWrapper[ID]): KeyWrapper[ID] = {
      if (reuse == null) {
        new KeyWrapper().setKey(data.srcIds(pos))
      } else {
        reuse.setKey(data.srcIds(pos))
      }
    }

    override def getKey(
      data: UncompressedInBlock[ID],
      pos: Int): KeyWrapper[ID] = {
      getKey(data, pos, null)
    }

    private def swapElements[@specialized(Int, Float) T](
      data: Array[T],
      pos0: Int,
      pos1: Int): Unit = {
      val tmp = data(pos0)
      data(pos0) = data(pos1)
      data(pos1) = tmp
    }

    override def swap(data: UncompressedInBlock[ID], pos0: Int, pos1: Int): Unit = {
      swapElements(data.srcIds, pos0, pos1)
      swapElements(data.dstEncodedIndices, pos0, pos1)
      swapElements(data.ratings, pos0, pos1)
    }

    override def copyRange(
      src: UncompressedInBlock[ID],
      srcPos: Int,
      dst: UncompressedInBlock[ID],
      dstPos: Int,
      length: Int): Unit = {
      System.arraycopy(src.srcIds, srcPos, dst.srcIds, dstPos, length)
      System.arraycopy(src.dstEncodedIndices, srcPos, dst.dstEncodedIndices, dstPos, length)
      System.arraycopy(src.ratings, srcPos, dst.ratings, dstPos, length)
    }

    override def allocate(length: Int): UncompressedInBlock[ID] = {
      new UncompressedInBlock(
        new Array[ID](length), new Array[Int](length), new Array[Float](length))
    }

    override def copyElement(
      src: UncompressedInBlock[ID],
      srcPos: Int,
      dst: UncompressedInBlock[ID],
      dstPos: Int): Unit = {
      dst.srcIds(dstPos) = src.srcIds(srcPos)
      dst.dstEncodedIndices(dstPos) = src.dstEncodedIndices(srcPos)
      dst.ratings(dstPos) = src.ratings(srcPos)
    }
  }

  /**
   * Creates in-blocks and out-blocks from rating blocks.
   * @param prefix prefix for in/out-block names
   * @param ratingBlocks rating blocks
   * @param srcPart partitioner for src IDs
   * @param dstPart partitioner for dst IDs
   * @return (in-blocks, out-blocks)
   */
  private def makeBlocks[ID: ClassTag](
    prefix: String,
    ratingBlocks: RDD[((Int, Int), RatingBlock[ID])],
    srcPart: Partitioner,
    dstPart: Partitioner,
    storageLevel: StorageLevel)(
      implicit srcOrd: Ordering[ID]): (RDD[(Int, InBlock[ID])], RDD[(Int, OutBlock)]) = {
    val inBlocks = ratingBlocks.map {
      case ((srcBlockId, dstBlockId), RatingBlock(srcIds, dstIds, ratings)) =>
        // The implementation is a faster version of
        // val dstIdToLocalIndex = dstIds.toSet.toSeq.sorted.zipWithIndex.toMap
        val start = System.nanoTime()
        val dstIdSet = new OpenHashSet[ID](1 << 20)
        dstIds.foreach(dstIdSet.add)
        val sortedDstIds = new Array[ID](dstIdSet.size)
        var i = 0
        var pos = dstIdSet.nextPos(0)
        while (pos != -1) {
          sortedDstIds(i) = dstIdSet.getValue(pos)
          pos = dstIdSet.nextPos(pos + 1)
          i += 1
        }
        assert(i == dstIdSet.size)
        Sorting.quickSort(sortedDstIds)
        val dstIdToLocalIndex = new OpenHashMap[ID, Int](sortedDstIds.length)
        i = 0
        while (i < sortedDstIds.length) {
          dstIdToLocalIndex.update(sortedDstIds(i), i)
          i += 1
        }
        logDebug(
          "Converting to local indices took " + (System.nanoTime() - start) / 1e9 + " seconds.")
        val dstLocalIndices = dstIds.map(dstIdToLocalIndex.apply)
        (srcBlockId, (dstBlockId, srcIds, dstLocalIndices, ratings))
    }.groupByKey(new ALSPartitioner(srcPart.numPartitions))
      .mapValues { iter =>
        val builder =
          new UncompressedInBlockBuilder[ID](new LocalIndexEncoder(dstPart.numPartitions))
        iter.foreach {
          case (dstBlockId, srcIds, dstLocalIndices, ratings) =>
            builder.add(dstBlockId, srcIds, dstLocalIndices, ratings)
        }
        builder.build().compress()
      }.setName(prefix + "InBlocks")
      .persist(storageLevel)
    val outBlocks = inBlocks.mapValues {
      case InBlock(srcIds, dstPtrs, dstEncodedIndices, _) =>
        val encoder = new LocalIndexEncoder(dstPart.numPartitions)
        val activeIds = Array.fill(dstPart.numPartitions)(mutable.ArrayBuilder.make[Int])
        var i = 0
        val seen = new Array[Boolean](dstPart.numPartitions)
        while (i < srcIds.length) {
          var j = dstPtrs(i)
          ju.Arrays.fill(seen, false)
          while (j < dstPtrs(i + 1)) {
            val dstBlockId = encoder.blockId(dstEncodedIndices(j))
            if (!seen(dstBlockId)) {
              activeIds(dstBlockId) += i // add the local index in this out-block
              seen(dstBlockId) = true
            }
            j += 1
          }
          i += 1
        }
        activeIds.map { x =>
          x.result()
        }
    }.setName(prefix + "OutBlocks")
      .persist(storageLevel)
    (inBlocks, outBlocks)
  }

  /**
   * Compute dst factors by constructing and solving least square problems.
   *
   * @param srcFactorBlocks src factors
   * @param srcOutBlocks src out-blocks
   * @param dstInBlocks dst in-blocks
   * @param rank rank
   * @param regParam regularization constant
   * @param srcEncoder encoder for src local indices
   * @param implicitPrefs whether to use implicit preference
   * @param alpha the alpha constant in the implicit preference formulation
   * @param solver solver for least squares problems
   *
   * @return dst factors
   */
  private def computeFactors[ID](
    biases: RDD[(Int, Array[Float])],
    srcFactorBlocks: RDD[(Int, FactorBlock)],
    srcOutBlocks: RDD[(Int, OutBlock)],
    dstInBlocks: RDD[(Int, InBlock[ID])],
    rank: Int,
    regParam: Double,
    srcEncoder: LocalIndexEncoder,
    implicitPrefs: Boolean = false,
    alpha: Double = 1.0,
    solver: LeastSquaresNESolver): RDD[(Int, FactorBlock)] = {
    val numSrcBlocks = srcFactorBlocks.partitions.length
    //  newSrcFactors.persist(StorageLevel.MEMORY_AND_DISK_SER)
    //    val YtY = if (implicitPrefs) Some(computeYtY(srcFactorBlocks, rank)) else None
    val tmp = srcFactorBlocks.map(x => {
      val f = x._2.map(v => {
        val t = v.clone()
        t(0) = 1.0f
        t
      })
      (x._1, f)
    })

    tmp.cache()

    val YtY = if (implicitPrefs) Some(computeYtY(tmp, rank + 1)) else None

    tmp.unpersist()

    val srcOut = srcOutBlocks.join(srcFactorBlocks)
      .flatMap {
        case (srcBlockId, (srcOutBlock, srcFactors)) =>
          srcOutBlock.view.zipWithIndex.map {
            case (activeIndices, dstBlockId) =>
              (dstBlockId, (srcBlockId, activeIndices.map(idx => srcFactors(idx))))
          }
        //            .map(x => ((x._1, x._2._1), x._2._2))
      }
    //      .map(x => (x._1._1, (x._1._2, x._2._1, x._2._2)))

    val merged = srcOut.groupByKey(new ALSPartitioner(dstInBlocks.partitions.length))
    val out = dstInBlocks
      .join(merged)
      .join(biases)
      .mapValues(x => (x._1._1, x._1._2, x._2))
      .mapValues {
        //x: (Int, (Iterable[ALS2.InBlock[ID]], Iterable[Iterable[(Int, Array[Array[Float]])]], Iterable[Array[Float]]))
        //((ALS2.InBlock[ID], Iterable[(Int, Array[Array[Float]])]), Array[Float]))
        case (InBlock(dstIds, srcPtrs, srcEncodedIndices, ratings), srcFactors, biases) =>
          val sortedSrcFactors = new Array[FactorBlock](numSrcBlocks)
          srcFactors.foreach {
            case (srcBlockId, factors) =>
              sortedSrcFactors(srcBlockId) = factors
          }
          val dstFactors = new Array[Array[Float]](dstIds.length)
          var j = 0
          val ls = new NormalEquation(rank + 1)
          while (j < dstIds.length) {
            ls.reset()
            if (implicitPrefs) {
              ls.merge(YtY.get)
            }
            var i = srcPtrs(j)
            var numExplicits = 0
            while (i < srcPtrs(j + 1)) {
              val encoded = srcEncodedIndices(i)
              val blockId = srcEncoder.blockId(encoded)
              val localIndex = srcEncoder.localIndex(encoded)
              val srcFactor = sortedSrcFactors(blockId)(localIndex)
              srcFactor(0) = 1.0f
              val bias = biases(j)
              val rating = ratings(i) - bias
              if (implicitPrefs) {
                // Extension to the original paper to handle b < 0. confidence is a function of |b|
                // instead so that it is never negative. c1 is confidence - 1.0.
                val c1 = alpha * math.abs(rating)
                // For rating <= 0, the corresponding preference is 0. So the term below is only added
                // for rating > 0. Because YtY is already added, we need to adjust the scaling here.
                if (rating > 0) {
                  numExplicits += 1
                  ls.add(srcFactor, (c1 + 1.0) / c1, c1)
                }
              } else {
                ls.add(srcFactor, rating)
                numExplicits += 1
              }
              i += 1
            }
            // Weight lambda by the number of explicit ratings based on the ALS-WR paper.
            dstFactors(j) = solver.solve(ls, numExplicits * regParam)
            j += 1
          }
          dstFactors
      }

    //restore factors
    out
  }

  /**
   * Computes the Gramian matrix of user or item factors, which is only used in implicit preference.
   * Caching of the input factors is handled in [[ALS#train]].
   */
  private def computeYtY(factorBlocks: RDD[(Int, FactorBlock)], rank: Int): NormalEquation = {
    factorBlocks.values.aggregate(new NormalEquation(rank))(
      seqOp = (ne, factors) => {
        factors.foreach(ne.add(_, 0.0))
        ne
      },
      combOp = (ne1, ne2) => ne1.merge(ne2))
  }

  /**
   * Encoder for storing (blockId, localIndex) into a single integer.
   *
   * We use the leading bits (including the sign bit) to store the block id and the rest to store
   * the local index. This is based on the assumption that users/items are approximately evenly
   * partitioned. With this assumption, we should be able to encode two billion distinct values.
   *
   * @param numBlocks number of blocks
   */
  private[recommendation] class LocalIndexEncoder(numBlocks: Int) extends Serializable {

    require(numBlocks > 0, s"numBlocks must be positive but found $numBlocks.")

    private[this] final val numLocalIndexBits =
      math.min(java.lang.Integer.numberOfLeadingZeros(numBlocks - 1), 31)
    private[this] final val localIndexMask = (1 << numLocalIndexBits) - 1

    /** Encodes a (blockId, localIndex) into a single integer. */
    def encode(blockId: Int, localIndex: Int): Int = {
      require(blockId < numBlocks)
      require((localIndex & ~localIndexMask) == 0)
      (blockId << numLocalIndexBits) | localIndex
    }

    /** Gets the block id from an encoded index. */
    @inline
    def blockId(encoded: Int): Int = {
      encoded >>> numLocalIndexBits
    }

    /** Gets the local index from an encoded index. */
    @inline
    def localIndex(encoded: Int): Int = {
      encoded & localIndexMask
    }
  }

  /**
   * Partitioner used by ALS. We requires that getPartition is a projection. That is, for any key k,
   * we have getPartition(getPartition(k)) = getPartition(k). Since the the default HashPartitioner
   * satisfies this requirement, we simply use a type alias here.
   */
  private[recommendation]type ALSPartitioner = org.apache.spark.HashPartitioner
}