StreamingSampling.jl

Documentation for StreamingSampling.

StreamingSampling is a Julia-based proof-of-concept implementation of a streamed variant of maximum-entropy sampling (UPmaxentropy). It is designed to process large datasets stored on disk with minimal impact on RAM. The method begins by computing first-order inclusion probabilities using a DPP-based heuristic, and then feeds these probabilities into the classical UPmaxentropy algorithm to produce diverse samples.