Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/3435#discussion_r20843304
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/feature/StandardScaler.scala ---
@@ -97,30 +97,51 @@ 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
+ val localShift = shift
+ vector match {
+ case dv: DenseVector =>
+ val values = dv.values.clone()
+ val size = values.size
var i = 0
- while (i < output.length) {
- output(i) = (output(i) - mean(i)) * (if (withStd) factor(i)
else 1.0)
- i += 1
+ if (withStd) {
+ val localFactor = factor
+ while (i < size) {
+ values(i) = (values(i) - localShift(i)) * localFactor(i)
+ i += 1
+ }
+ } else {
+ while (i < size) {
+ values(i) -= localShift(i)
+ i += 1
+ }
}
- Vectors.fromBreeze(output)
+ Vectors.dense(values)
case v => throw new IllegalArgumentException("Do not support
vector type " + v.getClass)
}
} else if (withStd) {
- vector.toBreeze match {
- case dv: BDV[Double] => Vectors.fromBreeze(dv :* factor)
- case sv: BSV[Double] =>
+ val localFactor = factor
+ vector match {
+ case dv: DenseVector =>
+ val values = dv.values.clone()
+ val size = values.size
+ var i = 0
+ while(i < size) {
+ values(i) *= localFactor(i)
+ i += 1
+ }
+ Vectors.dense(values)
+ case sv: SparseVector =>
// For sparse vector, the `index` array inside sparse vector
object will not be changed,
// so we can re-use it to save memory.
- val output = new BSV[Double](sv.index, sv.data.clone(),
sv.length)
+ val indices = sv.indices
+ val values = sv.values.clone()
+ val size = values.size
--- End diff --
`size` -> `nnz`
---
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]
