Uploaded image for project: 'Spark'
  1. Spark
  2. SPARK-3066

Support recommendAll in matrix factorization model

    XMLWordPrintableJSON

Details

    • New Feature
    • Status: Resolved
    • Major
    • Resolution: Fixed
    • None
    • 1.4.0
    • MLlib
    • None

    Description

      ALS returns a matrix factorization model, which we can use to predict ratings for individual queries as well as small batches. In practice, users may want to compute top-k recommendations offline for all users. It is very expensive but a common problem. We can do some optimization like

      1) collect one side (either user or product) and broadcast it as a matrix
      2) use level-3 BLAS to compute inner products
      3) use Utils.takeOrdered to find top-k

      Attachments

        Issue Links

          is duplicated by

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-4231 Add RankingMetrics to examples.MovieLensALS

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

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-4823 rowSimilarities

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

          Improvement - An improvement or enhancement to an existing feature or task. SPARK-11968 ALS recommend all methods spend most of time in GC

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

          Pull request #3098 [Github] Pull Request #3098 (debasish83)

          Pull request #5829 [Github] Pull Request #5829 (mengxr)

          Activity

            People

              debasish83 Debasish Das
              mengxr Xiangrui Meng
              Votes:
              0 Vote for this issue
              Watchers:
              6 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: