Details

    • Type: Bug
    • Status: Resolved
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 4.0.0
    • Component/s: Spark
    • Labels:
      None
    • Release Notes:
      Fixed a bug in the KMeans example that caused it to calculate the wrong cluster centroids
    • Rank:
      1|hzzm13:

      Description

      I tried running the Spark KMeans example with the attached data, which clearly has 2 centers. After I run the flow, then run the spark program, then query the service, I see centers:

      "210.45999999999995,210.45999999999995,109.42000000000002"
      "10.46,10.46,109.42000000000002"

      Clearly something is not right, as the y coordinate is exactly the same as the x, and not close to any of the y coordinates in the training data.

        Attachments

          Activity

            People

            • Assignee:
              ashau Albert Shau
              Reporter:
              ashau Albert Shau
            • Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: