srowen commented on a change in pull request #24963: [SPARK-28159][ML] Make the
transform natively in ml framework to avoid extra conversion
URL: https://github.com/apache/spark/pull/24963#discussion_r299758533
##########
File path:
mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAOptimizer.scala
##########
@@ -642,6 +639,20 @@ private[clustering] object OnlineLDAOptimizer {
}
val sstatsd = expElogthetad.asDenseMatrix.t * (ctsVector /:/
phiNorm).asDenseMatrix
- (gammad, sstatsd, ids)
+ (gammad, sstatsd, indices)
+ }
+
+ private[clustering] def variationalTopicInference(
+ termCounts: Vector,
+ expElogbeta: BDM[Double],
+ alpha: breeze.linalg.Vector[Double],
+ gammaShape: Double,
+ k: Int,
+ seed: Long): (BDV[Double], BDM[Double], List[Int]) = {
+ val (ids: List[Int], cts: Array[Double]) = termCounts match {
+ case v: DenseVector => ((0 until v.size).toList, v.values)
Review comment:
Looks good then except we might be able to make one more optimization here?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
