mboehm7 commented on pull request #1171: URL: https://github.com/apache/systemds/pull/1171#issuecomment-770117962
LGTM - thanks for the patch @OlgaOvcharenko, this is awesome and a great improvement for both, the spark and federated backends. During the merge I made some changes through to avoid unnecessary overhead and work correctly with binning: * Removed reflection, unnecessary type handling, and moved the object creation into the respective factory classes. * Fixed the binning serialization/deserialization to allow a variable number of bins per encoded column, and work in scenarios where the bin boundaries are not yet allocated. * Removed unused serialization test, and test data set. * Fixed warnings (e.g., type unsafe iteration over hash maps, unnecessary casts) ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
