spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sean Owen (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (SPARK-15399) Wrong equation is used in the method of org.apache.spark.mllib.clustering.KMeans
Date Thu, 19 May 2016 09:06:12 GMT

     [ https://issues.apache.org/jira/browse/SPARK-15399?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Sean Owen updated SPARK-15399:
------------------------------
               Flags:   (was: Patch,Important)
    Target Version/s:   (was: 1.6.1)
              Labels:   (was: patch)

> Wrong equation is used in the method of org.apache.spark.mllib.clustering.KMeans
> --------------------------------------------------------------------------------
>
>                 Key: SPARK-15399
>                 URL: https://issues.apache.org/jira/browse/SPARK-15399
>             Project: Spark
>          Issue Type: Bug
>          Components: MLlib
>    Affects Versions: 1.6.1
>         Environment: windows 64bit
>            Reporter: 雷文昌
>
> the method is in org.apache.spark.mllib.clustering.KMeans
> the equation |a-b|=||a|-|b|| is wrong when a and b are vector. but it is used in the
spark-1.6.1.
> private[mllib] def findClosest(
>       centers: TraversableOnce[VectorWithNorm],
>       point: VectorWithNorm): (Int, Double) = {
>     var bestDistance = Double.PositiveInfinity
>     var bestIndex = 0
>     var i = 0
>     centers.foreach { center =>
>       // Since `\|a - b\| \geq |\|a\| - \|b\||`, we can use this lower bound to avoid
unnecessary
>       // distance computation.
>       var lowerBoundOfSqDist = center.norm - point.norm
>       lowerBoundOfSqDist = lowerBoundOfSqDist * lowerBoundOfSqDist
>       if (lowerBoundOfSqDist < bestDistance) {
>         val distance: Double = fastSquaredDistance(center, point)
>         if (distance < bestDistance) {
>           bestDistance = distance
>           bestIndex = i
>         }
>       }
>       i += 1
>     }
>     (bestIndex, bestDistance)
>   }
> the center and the point in the source code are vector. and I suggest the code is that
> private[mllib] def findClosest(
>       centers: TraversableOnce[VectorWithNorm],
>       point: VectorWithNorm): (Int, Double) = {
>     var bestDistance = Double.PositiveInfinity
>     var bestIndex = 0
>     var i = 0
>     centers.foreach { center =>
>       // distance computation.
>       val distance: Double = fastSquaredDistance(center, point)
>       if (distance < bestDistance) {
>         bestDistance = distance
>         bestIndex = i
>       }
>       i += 1
>     }
>     (bestIndex, bestDistance)
>   }



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org


Mime
View raw message