org.apache.spark.ml.recommendation

Class ALS.UncompressedInBlock<ID>

Object

org.apache.spark.ml.recommendation.ALS.UncompressedInBlock<ID>

Enclosing class:

ALS

public static class ALS.UncompressedInBlock<ID>
extends Object

A block of (srcId, dstEncodedIndex, rating) tuples stored in primitive arrays.

Constructor Summary

Constructors

Constructor and Description
ALS.UncompressedInBlock(Object srcIds, int[] dstEncodedIndices, float[] ratings, scala.reflect.ClassTag<ID> evidence$7, scala.math.Ordering<ID> ord)

Method Summary

Methods

Modifier and Type	Method and Description
ALS.InBlock<ID>	compress() Compresses the block into an ALS.InBlock.
int[]	dstEncodedIndices()
int	length() Size the of block.
float[]	ratings()
Object	srcIds()

Methods inherited from class Object

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

Constructor Detail

ALS.UncompressedInBlock

public ALS.UncompressedInBlock(Object srcIds,
                       int[] dstEncodedIndices,
                       float[] ratings,
                       scala.reflect.ClassTag<ID> evidence$7,
                       scala.math.Ordering<ID> ord)

Method Detail

srcIds

public Object srcIds()

dstEncodedIndices

public int[] dstEncodedIndices()

ratings

public float[] ratings()

length

public int length()

Size the of block.

compress

public ALS.InBlock<ID> compress()

Compresses the block into an ALS.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.