Skip to content

fix: Support string IDs in GraphFrame powerIterationClustering #773

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

Merged
SemyonSinchenko merged 1 commit into from
Jan 20, 2026
Merged

fix: Support string IDs in GraphFrame powerIterationClustering #773

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
32 changes: 29 additions & 3 deletions core/src/main/scala/org/graphframes/GraphFrame.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -689,14 +689,40 @@ class GraphFrame private (
* @return
*/
def powerIterationClustering(k: Int, maxIter: Int, weightCol: Option[String]): DataFrame = {
val integralTypeEdges = if (hasIntegralIdType) {
edges
} else {
val pureIds =
indexedEdges.drop(SRC, DST).withColumnsRenamed(Map(LONG_SRC -> SRC, LONG_DST -> DST))
if (weightCol.isDefined) {
pureIds.select(
col(SRC),
col(DST),
col("attr").getField(weightCol.get).alias(weightCol.get))
} else {
pureIds
}
}
val powerIterationClustering =
new PowerIterationClustering().setK(k).setMaxIter(maxIter).setDstCol(DST).setSrcCol(SRC)
weightCol match {
case Some(col) => powerIterationClustering.setWeightCol(col).assignClusters(edges)
val result = weightCol match {
case Some(col) =>
powerIterationClustering.setWeightCol(col).assignClusters(integralTypeEdges)
case None =>
powerIterationClustering
.setWeightCol("_weight")
.assignClusters(edges.withColumn("_weight", lit(1.0)))
.assignClusters(integralTypeEdges.withColumn("_weight", lit(1.0)))
}

if (hasIntegralIdType) {
result
} else {
result
.join(
indexedVertices.select(col(LONG_ID).alias(ID), col(ID).alias("_ID")),
Seq(ID),
"inner")
.select(col("_ID").alias(ID), col("cluster"))
}
}

Expand Down
33 changes: 33 additions & 0 deletions core/src/test/scala/org/graphframes/GraphFrameSuite.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -588,6 +588,39 @@ class GraphFrameSuite extends SparkFunSuite with GraphFrameTestSparkContext {
assert(Seq(0, 0, 0, 0, 1, 0) == clusters)
}

test("power iteration clustering string ids") {
val spark = this.spark
import spark.implicits._
val edges = spark
.createDataFrame(
Seq(
("1", "0", 0.5),
("2", "0", 0.5),
("2", "1", 0.7),
("3", "0", 0.5),
("3", "1", 0.7),
("3", "2", 0.9),
("4", "0", 0.5),
("4", "1", 0.7),
("4", "2", 0.9),
("4", "3", 1.1),
("5", "0", 0.5),
("5", "1", 0.7),
("5", "2", 0.9),
("5", "3", 1.1),
("5", "4", 1.3)))
.toDF("src", "dst", "weight")
val vertices = Seq("0", "1", "2", "3", "4", "5").toDF("id")
val gf = GraphFrame(vertices, edges)
val clusters = gf
.powerIterationClustering(k = 2, maxIter = 40, weightCol = Some("weight"))
.collect()
.sortBy(_.getAs[String]("id"))
.map(_.getAs[Int]("cluster"))
.toSeq
assert(Seq(1, 1, 1, 1, 1, 0) == clusters)
}

test("convert directed graph to undirected") {
val v = spark.createDataFrame(Seq((1L, "a"), (2L, "b"), (3L, "c"))).toDF("id", "name")
val e = spark.createDataFrame(Seq((1L, 2L), (2L, 3L))).toDF("src", "dst")
Expand Down