How to apply Particle Swarm Optimization using SKO.pso python

Question

I find a package on the Github, it contains various algorithms about evolution. The link is below: https://github.com/guofei9987/scikit-opt

However, I got confused when using the package. When I try to do as the example, setting the dim as one, and lower bound as zero when upper bound is ten, focusing on the non negative for x, it got wrong answer, here is my code:

def demo_func(x):
    # Sphere
    x1= x
    return ((10-4*x1)/(4*x1+3))*x1
from sko.PSO import PSO
pso = PSO(func=demo_func, n_dim=1, pop=40, max_iter=150, lb=[0], ub=[10], w=0.8, c1=0.5, c2=0.5)
pso.run()
print('best_x is ', pso.gbest_x, 'best_y is', pso.gbest_y)

import matplotlib.pyplot as plt

plt.plot(pso.gbest_y_hist)
plt.show()

the result are below:result

best_x is [10.] best_y is [-6.97674419]

However, it is a wrong answer. the right answer for x is between 0.5 to 1. Does anybody has suggestions?

Answer 1

PSO optimize the minimum value of function. add a negative sign to get the max.

def demo_func(x):
    x1,= x
    return -((10-4*x1)/(4*x1+3))*x1
from sko.PSO import PSO
pso = PSO(func=demo_func, n_dim=1, pop=40, max_iter=150, lb=[0], ub=[10], w=0.8, c1=0.5, c2=0.5)
pso.run()
print('best_x is ', pso.gbest_x, 'best_y is', pso.gbest_y)

import matplotlib.pyplot as plt

plt.plot(pso.gbest_y_hist)
plt.show()