Mlpack provides an extensible, flexible exact nearest neighbor search implementation, with the ability to do dual-tree k-NN search. However, there is currently no support for approximate nearest neighbor search, despite its popularity. The aim of this project is the implementation of approximate K-NN search and then, work with the benchmarking system in order to produce rigorous comparisons between mlpack's approximate k-NN search implementation and other libraries, such as FLANN, ANN, or LSHKIT. Follow the progress of the project on:


Marcos Pividori


  • sumedhghaisas