Skip to content

[SPARK-58550][ML] Delay GaussianMixture aggregation allocations #57750

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
zhengruifeng wants to merge 1 commit into
base: master
Choose a base branch
from
+17 −5
Draft

[SPARK-58550][ML] Delay GaussianMixture aggregation allocations #57750

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 15 additions & 3 deletions mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -201,10 +201,22 @@ class GaussianMixture private (
val compute = sc.broadcast(ExpectationSum.add(weights, gaussians)_)

// aggregate the cluster contribution for all sample points
// Avoid allocating and serializing a large zero value for empty partitions.
val sums = breezeData.treeAggregate[ExpectationSum](
zeroValue = ExpectationSum.zero(k, d),
seqOp = (agg: ExpectationSum, v: BV[Double]) => compute.value(agg, v),
combOp = (agg1: ExpectationSum, agg2: ExpectationSum) => agg1 += agg2,
zeroValue = null.asInstanceOf[ExpectationSum],
seqOp = (maybeAgg: ExpectationSum, v: BV[Double]) => {
val agg = if (maybeAgg == null) ExpectationSum.zero(k, d) else maybeAgg
compute.value(agg, v)
},
combOp = (agg1: ExpectationSum, agg2: ExpectationSum) => {
if (agg1 == null) {
agg2
} else if (agg2 == null) {
agg1
} else {
agg1 += agg2
}
},
depth = 2,
finalAggregateOnExecutor = true)

Expand Down
4 changes: 2 additions & 2 deletions mllib/src/test/scala/org/apache/spark/mllib/clustering/GaussianMixtureSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,12 +40,12 @@ class GaussianMixtureSuite extends SparkFunSuite with MLlibTestSparkContext {
}
}

test("single cluster") {
test("single cluster with empty partitions") {
val data = sc.parallelize(Seq(
Vectors.dense(6.0, 9.0),
Vectors.dense(5.0, 10.0),
Vectors.dense(4.0, 11.0)
))
), 4)

// expectations
val Ew = 1.0
Expand Down