Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/3636#discussion_r26540822
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/optimization/GradientDescent.scala
---
@@ -219,4 +265,39 @@ object GradientDescent extends Logging {
(weights, stochasticLossHistory.toArray)
}
+
+ def runMiniBatchSGD(
+ data: RDD[(Double, Vector)],
+ gradient: Gradient,
+ updater: Updater,
+ stepSize: Double,
+ numIterations: Int,
+ regParam: Double,
+ miniBatchFraction: Double,
+ initialWeights: Vector): (Vector, Array[Double]) =
+ GradientDescent.runMiniBatchSGD(data, gradient, updater, stepSize,
numIterations,
+ regParam, miniBatchFraction,
initialWeights, 0.001)
+
+
+ private def isConverged(previousWeights: Vector, currentWeights: Vector,
+ initialWeights: Vector, convergenceTol: Double):
Boolean = {
+ require(previousWeights != None)
+ require(currentWeights != None)
+ // To compare with convergence tolerance
+ def solutionVecDiff(previousWeight: Vector,
+ currentWeight: Vector): Double = {
+
+ val lastWeight = currentWeight.toBreeze
+ val lastBeforeWeight = previousWeight.toBreeze
+ sum((lastBeforeWeight - lastWeight)
+ :* (lastBeforeWeight - lastWeight)) / lastWeight.length
+ }
+
+ def squareAvg(weights: Vector): Double =
+ sum(weights.toBreeze :* weights.toBreeze) / weights.toBreeze.length
+
+ solutionVecDiff(previousWeights, currentWeights) <
+ convergenceTol * squareAvg(initialWeights)
--- End diff --
I believe Breeze has norm() functions you can use instead.
This norm should be precomputed and stored.
---
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]
