This repository includes Julia codes for symbolic regression. The below instruction guides you how to install the requirements and run the code.
Install Julia. Available at Julia Download. This code is compatible with Julia 1.5.3.
Install SCIPOptSuite. Available at SCIPOptSuite Download. Before you download, please check which version is supported by the SCIP package in Julia SCIP.jl News. The most recent version SCIP v.7.0.2 is supported (confirmed on 02/04/2021).
Execute install_pkgs.jl to install the required julia packages.
example/data.txt is a sample dataset. The formula is x1*x2+1. Each row represents a data point. The code assumes that the last column is the dependent varaible.
main_example.jl is an example code to run MINLP and STreCH.
Once MINLP or STreCH is finished, it will create df_sol.csv, which includes the list of formulas and their objective values (MSE).
The paper is working at Overleaf.
data/ contains all the datasets used in the paper. Each folder in data/ includes a info file info.txt, two training set with/without noise, one validation set, and one testing set.
Notice: This software was developed in the course of or under prime
contract No. DE-AC02-06CH11357 between the U.S.
Department of Energy and UChicago Argonne, LLC.
This source code is licensed under the BSD-style license found in the LICENSE file in the root directory of this source tree.