tobiasrieger opened a new pull request #1154: URL: https://github.com/apache/systemds/pull/1154
This PR contains a lot of valuable features for the parameter server ### Validation There is now a second createPS function in the ParamservBuiltinCPInstruction. If all the additional arguments are specified the parameter server is able to validate after each epoch. It will do so if LOG.info is enabled. This feature is implemented for the federated parameter server ONLY, but is easily implemented for the other cases too. ### Federated Parameter Server Statistics The federated parameter server now uses the same statistics as the regular one, where possible. Also a number of new ones were introduced: * Aggregated Validation Time * Aggregated Fed Communication Time * Federated Data Partitioning Time * Aggregated Fed Batch Weighing Time * Fed Worker Computation Time (This includes gradient calculation, local updates and batch slicing. The granularity can be improved if needed.) ### Other Changes * scaleAndPushGradients was refactored to weighAndPushGradients * Logging cleanup * Not implemented exceptions * Validation Functions for the Federated Parameter Server Test DML files ---------------------------------------------------------------- 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]
