https://github.com/jbytecode/mccga.py

Machine-coded compact genetic algorithm in Python
https://github.com/jbytecode/mccga.py

Last synced: 22 days ago
JSON representation

Machine-coded compact genetic algorithm in Python

• Host: GitHub
• URL: https://github.com/jbytecode/mccga.py
• Owner: jbytecode
• License: mit
• Created: 2023-09-26T09:44:34.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-10-01T17:57:59.000Z (about 1 year ago)
• Last Synced: 2024-10-15T16:56:54.129Z (2 months ago)
• Language: Python
• Size: 13.7 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# mccga.py

Machine-coded compact genetic algorithm in Python

## In-short

The package implements the Machine-coded compact genetic algorithm defined in 

Satman, M. H. & Akadal, E. (2020). Machine Coded Compact Genetic Algorithms for Real Parameter Optimization Problems . Alphanumeric Journal , 8 (1) , 43-58 . DOI: 10.17093/alphanumeric.576919 [Link](https://dergipark.org.tr/en/pub/alphanumeric/issue/55603/576919)

## Usage 

Suppose the optimization problem is 

$$

\min f(x, y) = \text{abs}(x - 3.14159265) + \text{abs}(y - \exp{1})

$$

then the MCCGA searches for the minimum using 

```python 

def f(xs: list[float]) -> float:

    return abs(xs[0] - 3.14159265) + abs(xs[1] - 2.71828)

rangemin = [-100.0, -100.0]

rangemax = [100.0, 100.0]

mutrate = 0.001

maxiter = 100000

result = optimizer.mccga(f, rangemin, rangemax, mutrate, maxiter)

```

## Other implementations

- Julia (https://github.com/jmejia8/Metaheuristics.jl)

- Rust (https://crates.io/crates/mccga)

- Java (https://github.com/jbytecode/mccga.java)

## Calling mccga() from R

Thanks to the `reticulate` package, the Python function `mccga()` can be called into R.

Here is the example:

```R

# The package for R & Python integration

library(reticulate)

# Loading the mccga library

source_python("optimizer.py")

# Defining the objective function 

f <- function(xs){

    val <- (xs[1] - 3.14159265)^2 + (xs[2] - 2.71828)^2

    return(val)

}

result <- mccga(f, c(-100, -100), c(100, 100), 0.0001, 100000)

> result

[1] 3.141595 2.718276

```