Metadata-Version: 2.1
Name: rils-rols
Version: 0.3
Summary: RILS-ROLS: Robust Symbolic Regression via Iterated Local Search and Ordinary Least Squares
Home-page: https://github.com/kartelj/rils-rols
Author: Aleksandar Kartelj, Marko Đukanović
Author-email: aleksandar.kartelj@gmail.com
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE

RILS-ROLS is metaheuristic-based framework to deal with problems of symbolic regression. 

All of its aspects (method description, empirical results, etc.) are explained in the paper named:
"RILS-ROLS: Robust Symbolic Regression via Iterated Local Search and Ordinary Least Squares" by Aleksandar Kartelj and Marko Djukanovic. 
This paper is currently under review in the Journal of Big Data, Springer. 

All RILS-ROLS resources can be found at https://github.com/kartelj/rils-rols

RILS-ROLS distribution is available as a pip package at https://pypi.org/project/rils-rols

Minimal working example can be seen bellow:
```python
from rils_rols.rils_rols import RILSROLSRegressor
from math import sin, log

regr = RILSROLSRegressor()

X = [[3, 4], [1, 2], [-10, 20], [10, 10], [100, 100], [22, 23]]
y = [sin(x1)+2.3*log(x2) for x1, x2 in X]

regr.fit(X, y)

print("Final model is "+str(regr.model))
```

