The interface shared by all quantile finders, no matter if they are exact or approximate.
Read-only equi-depth histogram for selectivity estimation.
A class to test the QuantileBin1D code.
Factory constructing exact and approximate quantile finders for both known and unknown N.
The approx. algorithms are primarily intended to help applications scale. When faced with a large data sequence, traditional methods either need very large memories or time consuming disk based sorting. In constrast, the approx. algorithms can deal with > 10^10 values without disk based sorting.
All classes can be seen from various angles, for example as
QuantileFinderFactoryand the interface
DoubleQuantileFinderto learn more. Most users will never need to know more than how to use these. Actual implementations of the QuantileFinder interface are hidden. They are indirectly constructed via the the factory.
hep.aida.bin.QuantileBin1D, demonstrating how this package can be used.
Copyright © 2006–2016 SYSTAP, LLC DBA Blazegraph. All rights reserved.