# basicGA

**Repository Path**: hnytzy/basicGA

## Basic Information

- **Project Name**: basicGA
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2024-08-13
- **Last Updated**: 2024-08-13

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Practise Genetic Algorithm in Python [![Build Status](https://travis-ci.com/dothinking/basicGA.svg?branch=master)](https://travis-ci.com/dothinking/basicGA)

- single-objective minimization
- built-in Individual and Operators for mathematical functions, travelling salesman problem
- user-defined Individuals and Operators
- based on Numpy

An example for solving the minimum solution of common test function `schaffer N4` is shown below.

## Import modules

```python
from GA.GAPopulation.DecimalIndividual import DecimalFloatIndividual
from GA.GAPopulation.Population import Population
from GA.GAOperators.Selection import RouletteWheelSelection
from GA.GAOperators.Crossover import DecimalCrossover
from GA.GAOperators.Mutation import DecimalMutation
from GA.GAProcess import GA
import numpy as np
```

## Initialize GA

```python
# individual
dimension = 2
ranges = [(-10, 10)] * dimension
I = DecimalFloatIndividual(ranges)

# population
P = Population(I, 50)

# operators
S = RouletteWheelSelection()
C = DecimalCrossover([0.5, 0.9], 0.5) # adaptive crossover rate
# C = DecimalCrossover(0.9, 0.5) # constant crossover rate
M = DecimalMutation(0.12)

# GA
g = GA(P, S, C, M)
```

## Solve

```python
# schaffer-N4
# sol: x=[0,1.25313], min=0.292579
schaffer_n4 = lambda x: 0.5 + (np.cos(np.sin(abs(x[0]**2-x[1]**2)))**2-0.5) / (1.0+0.001*(x[0]**2+x[1]**2))**2

res = g.run(schaffer_n4, gen=800)   

x = [0,1.25313]
print('{0} : {1}'.format(res.evaluation, res.solution)) # 0.29257882535592317 : [1.25339239e+00 6.28576519e-05]
print('error: {:<3f} %'.format((res.evaluation/schaffer_n4(x)-1.0)*100)) # error: 0.000066 %
```