Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-21926

Compatibility between ML Transformers and Structured Streaming

    XMLWordPrintableJSON

Details

    • Umbrella
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • 2.2.0
    • 2.3.0
    • ML, Structured Streaming
    • None

    Description

      We've run into a few cases where ML components don't play nice with streaming dataframes (for prediction). This ticket is meant to help aggregate these known cases in one place and provide a place to discuss possible fixes.

      Failing cases:
      1) VectorAssembler where one of the inputs is a VectorUDT column with no metadata.
      Possible fixes:
      More details here SPARK-22346.

      2) OneHotEncoder where the input is a column with no metadata.
      Possible fixes:
      a) Make OneHotEncoder an estimator (SPARK-13030).
      b) Allow user to set the cardinality of OneHotEncoder.

      Attachments

        Issue Links

          contains

          Bug - A problem which impairs or prevents the functions of the product. SPARK-22888 OneVsRestModel does not work with Structured Streaming

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          New Feature - A new feature of the product, which has yet to be developed. SPARK-24465 LSHModel should support Structured Streaming for transform

          • Major - Major loss of function.
          • Resolved

          Test - A new unit, integration or system test. SPARK-22644 Make ML testsuite support StructuredStreaming test

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

          Bug - A problem which impairs or prevents the functions of the product. SPARK-23037 RFormula should not use deprecated OneHotEncoder and should include VectorSizeHint in pipeline

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-22346 Update VectorAssembler to work with Structured Streaming

          • Critical - Crashes, loss of data, severe memory leak.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-21748 Migrate the implementation of HashingTF from MLlib to ML

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

          Bug - A problem which impairs or prevents the functions of the product. SPARK-19141 VectorAssembler metadata causing memory issues

          • Major - Major loss of function.
          • Resolved

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-13030 Change OneHotEncoder to Estimator

          • Major - Major loss of function.
          • Resolved

          Documentation - Documentation or Website SPARK-22735 Add VectorSizeHint to ML features documentation

          • Major - Major loss of function.
          • Resolved

          Documentation - Documentation or Website SPARK-23048 Update mllib docs to replace OneHotEncoder with OneHotEncoderEstimator

          • Major - Major loss of function.
          • Resolved

          Activity

            People

              Unassigned Unassigned
              bago.amirbekian Bago Amirbekian
              Votes:
              3 Vote for this issue
              Watchers:
              16 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: