PipelineDP is a Python framework for applying differentially private aggregations to large datasets using batch processing systems such as Apache Spark, Apache Beam, and more.
This PR implements the computation of UtilityReport hisogram per partition size.
Histogram description:
Hisogram bucket bounds are [0, 1, 10, 20, 50, 100, 200, 500, 1000, ...].
Hisotgram contains UtilityReport (which is aggregated Utility analysis information) for partitions of corresponding size, e.g. it contains UtilityReport for partitions of size [10, 20), UtilityReport for partitions of size [20, 50) etc.
Implementation details:
1.UtilityReportHistogram dataclass with histogram [partition_size_from, partition_size_to] -> UtilityReport
Computing UtilityReportHistogram in perform_utility_analysis. That's done by extending the existing computation by computation histogram as well.
This PR implements the computation of
UtilityReporthisogram per partition size.Histogram description:
Hisogram bucket bounds are [0, 1, 10, 20, 50, 100, 200, 500, 1000, ...]. Hisotgram contains
UtilityReport(which is aggregated Utility analysis information) for partitions of corresponding size, e.g. it containsUtilityReportfor partitions of size [10, 20),UtilityReportfor partitions of size [20, 50) etc.Implementation details: 1.
UtilityReportHistogramdataclass with histogram[partition_size_from, partition_size_to]->UtilityReportUtilityReportHistograminperform_utility_analysis. That's done by extending the existing computation by computation histogram as well.