Crash on custom objective function

[ibell]

• PySwarms version: head
• Python version: 3.6
• Operating System: linux

Description

Tried to optimize a harder function than the sphere, which is basically trivial. The sphere function, with all else held constant, works fine.

What I Did

import pyswarms as ps
from numpy import cos, sqrt

def Griewangk(x):
    sum1 = 0 
    prod1 = 1
    for i in range(len(x)):
        sum1 += pow(x[i], 2)
        prod1 *= cos(x[i]/sqrt(i+1))
    f = sum1/4000.0 - prod1 + 1
    return f

# Set-up hyperparameters
options = {'c1': 0.5, 'c2': 0.3, 'w':0.9}

# Call instance of GlobalBestPSO
optimizer = ps.single.GlobalBestPSO(n_particles=100, dimensions=10,
                                    options=options)

# Perform optimization
stats = optimizer.optimize(Griewangk, iters=1000)
print(stats)

yields

ValueError                                Traceback (most recent call last)
<ipython-input-40-515ac38c5127> in <module>()
     19 
     20 # Perform optimization
---> 21 stats = optimizer.optimize(Griewangk, iters=1000)
     22 print(stats)

~/miniconda3/envs/jupy36/lib/python3.6/site-packages/pyswarms/single/global_best.py in optimize(self, objective_func, iters, print_step, verbose)
    153             _m = np.repeat(m[:, np.newaxis], self.dimensions, axis=1)
    154             self.personal_best_pos = np.where(~_m, self.personal_best_pos,
--> 155                                               self.pos)
    156 
    157             # Get theValueError                                Traceback (most recent call last)
<ipython-input-40-515ac38c5127> in <module>()
     19 
     20 # Perform optimization
---> 21 stats = optimizer.optimize(Griewangk, iters=1000)
     22 print(stats)

~/miniconda3/envs/jupy36/lib/python3.6/site-packages/pyswarms/single/global_best.py in optimize(self, objective_func, iters, print_step, verbose)
    153             _m = np.repeat(m[:, np.newaxis], self.dimensions, axis=1)
    154             self.personal_best_pos = np.where(~_m, self.personal_best_pos,
--> 155                                               self.pos)
    156 
    157             # Get the minima of the pbest and check if it's less than

ValueError: operands could not be broadcast together with shapes (10,10) (100,10) (100,10)  minima of the pbest and check if it's less than

ValueError: operands could not be broadcast together with shapes (10,10) (100,10) (100,10) 

[ljvmiranda921]

Hi @ibell , let me get to this during the weekend, currently traveling so I can't access my work laptop

[ibell]

Actually, the bug was/is in my understanding. I didn't realize that the inputs to the objective functions should be vector inputs (all inputs are evaluated in parallel). Maybe that just needs to be clarified in the docs. I think in the docs it makes more sense to use a hard-coded function so you can see what is going on rather than using a baked in function.

[ljvmiranda921]

Hi @ibell , understood. Yes you are correct, the documentation still leaves a lot for improvement and it would be nice to do these upgrades. Just in case, would you also like to include this 'clarification' in your Pull Request?

[ljvmiranda921]

Reference #107

[rfpg1]

I know this is quite old
But i bumped into the same problem and I can't figure out how to write my own function and pass it as objective function
Could you give me some instructions?
I'm trying to make a simple example:

import pyswarms as ps
from pyswarms.utils.functions import single_obj as fx

def f(x):
    return x**2

options = {'c1':0.5, 'c2': 0.3, 'w':0.9}

optimizer = ps.single.GlobalBestPSO(n_particles=100, dimensions=2, options=options)

best_cost, best_pos = optimizer.optimize(f, iters=100)

print(best_cost)
print(best_pos)