POETs.jl is a Julia package that implements the Pareto Optimal Ensemble Techniques (POETs) method for multiobjective optimization. For background on the original POETs algorithm see:
and for its updated implementation in Julia:
Within Julia, use the clone command of the package manager to download and install the POETs repository:
Pkg.clone("git://github.com/varnerlab/POETs.jl")
To use POETs in your project (following installation) simply issue the command:
using POETs
To test the POETs installation use:
Pkg.test("POETs")
which runs test examples from the test directory. Lastly, to delete the JuPOETs package use the command:
Pkg.rm("POETs")
First, make sure Julia is installed on your computer. Next, clone the JuPOETs package to your Julia installation (using the Pkg.clone() command shown above). Lastly, write your problem specific functions (objective,neighbor,cooling and acceptance) which are required by JuPOETs, and your ready to go. We've included sample code to help you get started with JuPOETs in your project. The sample can be found in the sample/biochemical subdirectory. The sample encodes the estimation of an ensemble of biochemical model parameters from four conflicting training data sets. The driver for this sample is given in the run_biochemical_test.jl file, while the objective,neighbor,cooling and acceptance functions (required by JuPOETs) are encoded in the hcmem_lib.jl library.