Github user dbtsai commented on a diff in the pull request:
https://github.com/apache/spark/pull/3462#discussion_r20961188
--- Diff: mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala
---
@@ -261,6 +261,57 @@ object Vectors {
sys.error("Unsupported Breeze vector type: " + v.getClass.getName)
}
}
+
+ /**
+ * Returns the p-norm of this vector.
+ * @param vector input vector.
+ * @param p norm.
+ * @return norm in L^p^ space.
+ */
+ private[spark] def norm(vector: Vector, p: Double): Double = {
+ require(p >= 1.0)
+ val values = vector match {
+ case dv: DenseVector => dv.values
+ case sv: SparseVector => sv.values
+ case v => throw new IllegalArgumentException("Do not support vector
type " + v.getClass)
+ }
+ val size = values.size
+
+ if (p == 1) {
+ var sum = 0.0
+ var i = 0
+ while (i < size) {
+ sum += math.abs(values(i))
+ i += 1
+ }
+ sum
+ } else if (p == 2) {
+ var sum = 0.0
+ var i = 0
+ while (i < size) {
+ sum += values(i) * values(i)
--- End diff --
They will generate different bytecode, but I don't see the performance
difference. Maybe `DSTORE` to store the `value` to local variable is more
expensive than looping up twice. Or maybe JVM just optimizes it internally. I
don't have preference.
```scala
sum += values(i) * values(i)
```
will translate to
```
L24
LINENUMBER 292 L24
DLOAD 13
ALOAD 4
ILOAD 15
DALOAD
ALOAD 4
ILOAD 15
DALOAD
DMUL
DADD
DSTORE 13
```
while
```scala
val value = values(i)
sum += value * value
```
will translate to
```
L24
LINENUMBER 292 L24
ALOAD 4
ILOAD 15
DALOAD
DSTORE 16
L25
LINENUMBER 293 L25
DLOAD 13
DLOAD 16
DLOAD 16
DMUL
DADD
DSTORE 13
```
---
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]
