Github user dbtsai commented on a diff in the pull request:
https://github.com/apache/spark/pull/3435#discussion_r20885451
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/feature/StandardScaler.scala ---
@@ -97,30 +97,57 @@ class StandardScalerModel private[mllib] (
override def transform(vector: Vector): Vector = {
require(mean.size == vector.size)
if (withMean) {
- vector.toBreeze match {
- case dv: BDV[Double] =>
- val output = vector.toBreeze.copy
- var i = 0
- while (i < output.length) {
- output(i) = (output(i) - mean(i)) * (if (withStd) factor(i)
else 1.0)
- i += 1
+ // By default, Scala generates Java methods for member variables. So
every time when
+ // the member variables are accessed, `invokespecial` will be called
which is expensive.
+ // This can be avoid by having a local reference of `shift`.
+ val localShift = shift
--- End diff --
For different implementation of vector, toArray can be very expensive. For
example, toArray for sparse vector requires to create a new array object and
loop through all the non zero values. As a result, we can have a global `lazy
shift` which can prevent this happens.
---
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.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
