Default gammaShape

Author: Feynman Liang

mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAModel.scala

@@ -27,7 +27,6 @@ import org.json4s.jackson.JsonMethods._
 import org.apache.spark.SparkContext
 import org.apache.spark.annotation.Experimental
 import org.apache.spark.api.java.JavaPairRDD
-import org.apache.spark.broadcast.Broadcast
 import org.apache.spark.graphx.{Edge, EdgeContext, Graph, VertexId}
 import org.apache.spark.mllib.linalg.{Matrices, Matrix, Vector, Vectors}
 import org.apache.spark.mllib.util.{Loader, Saveable}
@@ -190,7 +189,8 @@ class LocalLDAModel private[clustering] (
     val topics: Matrix,
     override val docConcentration: Vector,
     override val topicConcentration: Double,
-    override protected[clustering] val gammaShape: Double) extends LDAModel with Serializable {
+    override protected[clustering] val gammaShape: Double = 100
+  ) extends LDAModel with Serializable {
 
   override def k: Int = topics.numCols
 
@@ -455,8 +455,9 @@ class DistributedLDAModel private[clustering] (
     val vocabSize: Int,
     override val docConcentration: Vector,
     override val topicConcentration: Double,
-    override protected[clustering] val gammaShape: Double,
-    private[spark] val iterationTimes: Array[Double]) extends LDAModel {
+    private[spark] val iterationTimes: Array[Double],
+    override protected[clustering] val gammaShape: Double = 100
+  ) extends LDAModel {
 
   import LDA._
 

mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAOptimizer.scala

@@ -209,11 +209,11 @@ final class EMLDAOptimizer extends LDAOptimizer {
   override private[clustering] def getLDAModel(iterationTimes: Array[Double]): LDAModel = {
     require(graph != null, "graph is null, EMLDAOptimizer not initialized.")
     this.graphCheckpointer.deleteAllCheckpoints()
-    // This assumes gammaShape = 100 in OnlineLDAOptimizer to ensure equivalence in LDAModel.toLocal
-    // conversion
+    // The constructor's default arguments assume gammaShape = 100 to ensure equivalence in
+    // LDAModel.toLocal conversion
     new DistributedLDAModel(this.graph, this.globalTopicTotals, this.k, this.vocabSize,
       Vectors.dense(Array.fill(this.k)(this.docConcentration)), this.topicConcentration,
-      100, iterationTimes)
+      iterationTimes)
   }
 }