kmundnic / ICASSP2019

Code and data accompanying the paper "Bluetooth based indoor localization using Triplet Embeddings"

ICASSP2019

Code accompanying the paper "Bluetooth based indoor localization using Triplet Embeddings"

This code runs in Julia v0.6 (tested on 0.6.4). The following packages are needed (run the following code to install):

Pkg.add("CSV")
Pkg.add("SCS")
Pkg.add("JuMP")
Pkg.add("JSON")
Pkg.add("Plots") # Will need a backend
Pkg.add("PyCall")
Pkg.add("PyPlot")
Pkg.add("DataFrames")

Running the experiments

Enabling multithreading

TripletEmbeddings.jl implements tSTE using threads to compute the gradient. To allow Julia to use more than one thread, run in the terminal export JULIA_NUM_THREADS=n (where n is the number of threads you want to use) before starting the Julia REPL.

Experiments

To run, open a Julia REPL in the same directory and run:

include("experiments.jl")