Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-31032

GMM compute summary and update distributions in one pass

    XMLWordPrintableJSON

    Details

    • Type: Improvement
    • Status: Resolved
    • Priority: Minor
    • Resolution: Fixed
    • Affects Version/s: 3.1.0
    • Fix Version/s: 3.1.0
    • Component/s: ML
    • Labels:
      None

      Description

      In current impl, GMM need to trigger two jobs at one iteration:

      1, one to compute summary;

      2, if {{shouldDistributeGaussians}} ((k - 1.0) / k) * numFeatures > 25.0,

      trigger another to update distributions;


       

      shouldDistributeGaussians is almost true in practice, since numFeatures is likely to be greater than 25.


       

      We can use only one job to impl above computation,

        Attachments

          Issue Links

            Activity

              People

              • Assignee:
                podongfeng zhengruifeng
                Reporter:
                podongfeng zhengruifeng
              • Votes:
                0 Vote for this issue
                Watchers:
                1 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved: