Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-3261

KMeans clusterer can return duplicate cluster centers

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Minor
    • Resolution: Fixed
    • 1.0.2
    • 2.1.0
    • MLlib

    Description

      This is a bad design choice. I think that it is preferable to produce no duplicate cluster centers. So instead of forcing the number of clusters to be K, return at most K clusters.

      Attachments

        Issue Links

          is related to

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-17389 KMeans speedup with better choice of k-means|| init steps = 2

          • Minor - Minor loss of function, or other problem where easy workaround is present.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-18427 Update docs of mllib.KMeans

          • Trivial - Cosmetic problem like misspelt words or misaligned text.
          • Resolved
          relates to

          Bug - A problem which impairs or prevents the functions of the product. SPARK-19319 SparkR Kmeans summary returns error when the cluster size doesn't equal to k

          • Major - Major loss of function.
          • Resolved
          links to

          Pull request #2419 [Github] Pull Request #2419 (derrickburns)

          Pull request #2634 [Github] Pull Request #2634 (derrickburns)

          Pull request #15450 [Github] Pull Request #15450 (srowen)

          Activity

            People

              srowen Sean R. Owen
              derrickburns Derrick Burns
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: