Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-44909

Skip starting torch distributor log streaming server when it is not available

    XMLWordPrintableJSON

Details

    • Improvement
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 0.5.0
    • 3.5.0, 4.0.0
    • ML
    • None

    Description

      Skip starting torch distributor log streaming server when it is not available.

       

      In some cases, e.g., in a databricks connect cluster, there is some network limitation that casues starting log streaming server failure, but, this does not need to break torch distributor training routine.

      Attachments

        Activity

          People

            weichenxu123 Weichen Xu
            weichenxu123 Weichen Xu
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: