Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/407#discussion_r11694640
--- Diff:
mllib/src/test/scala/org/apache/spark/mllib/recommendation/ALSSuite.scala ---
@@ -128,6 +128,47 @@ class ALSSuite extends FunSuite with LocalSparkContext
{
assert(u11 != u2)
}
+ test("custom partitioner") {
+ testALS(50, 50, 2, 15, 0.7, 0.3, false, false, false, 3, null)
+ testALS(50, 50, 2, 15, 0.7, 0.3, false, false, false, 3, new
Partitioner {
+ def numPartitions(): Int = 3
+ def getPartition(x: Any): Int = x match {
+ case null => 0
+ case _ => x.hashCode % 2
+ }
+ })
+ }
+
+ test("negative ids") {
+ val data = ALSSuite.generateRatings(50, 50, 2, 0.7, false, false)
+ val ratings = data._1.toArray
+ val correct = data._2
+ for (i <- 0 until ratings.length) {
+ var u = ratings(i).user
+ var p = ratings(i).product
+ var r = ratings(i).rating
+ u = u - 25
+ p = p - 25
+ ratings(i) = new Rating(u,p,r)
+ }
+ val ratingsRDD = sc.parallelize(ratings)
+
+ val model = ALS.train(ratingsRDD, 5, 15)
+
+ var pairs = new Array[(Int, Int)](0)
+ for (u <- -25 until 25; p <- -25 until 25) {
+ pairs = pairs :+ (u, p)
+ }
+ val ans = model.predict(sc.parallelize(pairs)).collect
+ for (r <- ans) {
--- End diff --
Use `foreach` instead of `for`. The former is faster.
~~~
ans.foreach { r =>
...
}
~~~
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
