optymus 0.1.2

Optimization Methods Library

Check the Optimization in Deep Learning and Engineering material.

Optymus is part of quantsci project.

This library provides a comprehensive collection of optimization methods, both with and without constraints. The main goal is provide a simple structure to improve research and development in optimization problems.

Implemented Methods

Method	Description
bfgs	Broyden-Fletcher-Goldfarb-Shanno (BFGS)
steepdesc	Steepest Descent
newton_raphson	Newton-Raphson Method
powell	Powell's Method
fletcher_reeves	Fletcher-Reeves

Getting Started

To begin using optymus, follow these steps:

1. Install optymus:

   pip install optymus
   

2. Explore the Documentation: Visit the official documentation to understand the available optimization methods and how to use them effectively.

3. Get Started:

   from optymus.optim import Optimizer
   from optymus.utils import mccormick_function
   
   import numpy as np
   
   f = mccormick_function()
   initial_point = np.array([2, 2])
   
   opt = Optimizer(f_obj=f,
                   x0=initial_point,
                   method='bfgs')
   
   opt.report()
   
   opt.plot()
   

Refer to the documentation for detailed information on each method and its application.

Implement your own method an compare with the implemented ones

We are working to implement a simple way to add your own optimization method.

Contributions

Contributions to Optymus are highly appreciated. If you have additional optimization methods, improvements, or bug fixes, please submit a pull request following the contribution guidelines.

Cite

If you use Optymus in your research, please consider citing the library using the following BibTeX entry:

@misc{optymus2024,
  author = {Costa, Kleyton and Menezes, Ivan},
  title = {Optymus: Optimization Methods Library for Python},
  year = {2024},
  note = {GitHub Repository},
  url = {https://github.com/quantsci/optymus}
}