rxPredict in a Distributed Compute Context RRS feed

  • Question

  • In a local compute context, the prediction results of rxPredict are appended to the test/prediction XDF file. 

    It seems that rxPredict works in a different way in a Spark/Hadoop compute context. The test/prediction XDF file is overwritten by the prediction results.

    Is there any way to append predictions without deleting the original columns in a distributed context?

    Tuesday, December 20, 2016 8:26 AM