Similarity Search Indexes based on Neighborhood Approximation

Installation

using Pkg
Pkg.add("NeighborhoodApproximation")

Usage

using SimilaritySearch
using NeighborhoodApproximationIndex
function create_dataset(n, dim)
    [rand(Float32, dim) for i in 1:n]
end

X = create_dataset(10_000, 8)
queries = create_dataset(100, 8)
napp = fit(DeloneInvIndex, l2_distance, X; numcenters=128, initial=:rand, maxiters=3, region_expansion=3)

See examples at examples directory of this repository. These are Pluto.jl notebooks, and probably you must install it in you wanna take a look.

About the methods in this package

This package contains some implementations of approximate similarity search methods based on neighborhood approximation, similar to those found in

Edgar Chavez, Mario Graff, Gonzalo Navarro, Eric S. Tellez:
Near neighbor searching with K nearest references. Inf. Syst. 51: 43-61 (2015)

Eric S. Tellez, Edgar Chavez, Gonzalo Navarro: Succinct nearest neighbor search. Inf. Syst. 38(7): 1019-1030 (2013)

Eric S. Tellez, Edgar Chavez, Gonzalo Navarro: Succinct nearest neighbor search. SISAP 2011: 33-40 (2011)

I am rewritting some of this methods in Julia. Therefore, by now, you will not find a complete set of features (like compact datastructures) or a broad exploration in the distance functions in the mapped space. The precise implementations with compression can be found in the old C# library natix.