[SPARK-5536] replace old ALS implementation by the new one

The only issue is that `analyzeBlock` is removed, which was marked as a developer API. I didn't change other tests in the ALSSuite under `spark.mllib` to ensure that the implementation is correct.

CC: srowen coderxiang

Author: Xiangrui Meng <meng@databricks.com>

Closes #4321 from mengxr/SPARK-5536 and squashes the following commits:

5a3cee8 [Xiangrui Meng] update python tests that are too strict
e840acf [Xiangrui Meng] ignore scala style check for ALS.train
e9a721c [Xiangrui Meng] update mima excludes
9ee6a36 [Xiangrui Meng] merge master
9a8aeac [Xiangrui Meng] update tests
d8c3271 [Xiangrui Meng] remove analyzeBlocks
d68eee7 [Xiangrui Meng] add checkpoint to new ALS
22a56f8 [Xiangrui Meng] wrap old ALS
c387dff [Xiangrui Meng] support random seed
3bdf24b [Xiangrui Meng] make storage level configurable in the new ALS

2 files changed, 1 insertions, 19 deletions

diff --git a/mllib/src/test/scala/org/apache/spark/ml/recommendation/ALSSuite.scala b/mllib/src/test/scala/org/apache/spark/ml/recommendation/ALSSuite.scala
index ee08c3c327..acc447742b 100644
--- a/mllib/src/test/scala/org/apache/spark/ml/recommendation/ALSSuite.scala
+++ b/mllib/src/test/scala/org/apache/spark/ml/recommendation/ALSSuite.scala
@@ -414,7 +414,7 @@ class ALSSuite extends FunSuite with MLlibTestSparkContext with Logging {
     val (training, test) =
       genExplicitTestData(numUsers = 20, numItems = 40, rank = 2, noiseStd = 0.01)
     for ((numUserBlocks, numItemBlocks) <- Seq((1, 1), (1, 2), (2, 1), (2, 2))) {
-      testALS(training, test, maxIter = 4, rank = 2, regParam = 0.01, targetRMSE = 0.03,
+      testALS(training, test, maxIter = 4, rank = 3, regParam = 0.01, targetRMSE = 0.03,
         numUserBlocks = numUserBlocks, numItemBlocks = numItemBlocks)
     }
   }
diff --git a/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala b/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala
index e9fc37e000..8775c0ca9d 100644
--- a/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala
+++ b/mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala
@@ -24,9 +24,7 @@ import scala.util.Random
 import org.scalatest.FunSuite
 import org.jblas.DoubleMatrix
 
-import org.apache.spark.SparkContext._
 import org.apache.spark.mllib.util.MLlibTestSparkContext
-import org.apache.spark.mllib.recommendation.ALS.BlockStats
 import org.apache.spark.storage.StorageLevel
 
 object ALSSuite {
@@ -189,22 +187,6 @@ class ALSSuite extends FunSuite with MLlibTestSparkContext {
     testALS(100, 200, 2, 15, 0.7, 0.4, false, false, false, -1, -1, false)
   }
 
-  test("analyze one user block and one product block") {
-    val localRatings = Seq(
-      Rating(0, 100, 1.0),
-      Rating(0, 101, 2.0),
-      Rating(0, 102, 3.0),
-      Rating(1, 102, 4.0),
-      Rating(2, 103, 5.0))
-    val ratings = sc.makeRDD(localRatings, 2)
-    val stats = ALS.analyzeBlocks(ratings, 1, 1)
-    assert(stats.size === 2)
-    assert(stats(0) === BlockStats("user", 0, 3, 5, 4, 3))
-    assert(stats(1) === BlockStats("product", 0, 4, 5, 3, 4))
-  }
-
-  // TODO: add tests for analyzing multiple user/product blocks
-
   /**
    * Test if we can correctly factorize R = U * P where U and P are of known rank.
    *