User-provided initial positions

[szhan]

Is your feature request related to a problem? Please describe.
I would like to specify the starting positions of every particle in a swarm.

Describe the solution you'd like
I see that init_pos can be specified, but it only allows the user to specify a starting position for the swarm's center. Looking at the code of pyswarms.base.base_single, it seems that reset() needs to be extended or modified. That is, I need to update how self.pos is being initialized. So, I would provide a list of starting positions to self.pos. Is there an already existing alternative to this?

[ljvmiranda921]

Hi @szhan , currently there is no explicit implementation for that, but we're currently developing a feature for this in #115 and #117 . The only workaround we can have here is to initialize the swarm as usual, then set the pos attribute into the positions my_own_initial_pos you need (making sure that the shape of my_own_init_pos is the same as pos):

import numpy as np
import pyswarms as ps

# Your own initial positions of shape (3,3)
my_own_init_pos = np.array([[1,2,3], [4,5,6], [7,8,9]])

# Initialize optimizer, this creates a position and velocity matrix of shape (3,3)
options = {'c1':0.5, 'c2':0.4,'w':0.9} 
optimizer = ps.single.GlobalBestPSO(n_particles=3, dimensions=3, options=options)

# Provide your own init_pos
optimizer.pos = my_own_init_pos.copy()

# Do optimization below
# optimizer.optimizer()...

I realized that this functionality is actually important upon releasing, so hopefully the next update will give this capability. For now, I hope this workaround works.

[szhan]

Thanks, @ljvmiranda921! Also, I would like to help with the development of PySwarms, as I find it quite useful in my work.

[szhan]

I have tried the workaround you suggested, but I am encountering an error that I cannot make sense of. I wrote a wrapper to feed GlobalBestPSO() some initial positions to solve some BBOB function.

When trying to have it solve the Rastrigin function, I got the following error:

ValueError: Input for Rastrigin function must be within [-5.12, 5.12].

The initial positions are clearly within bounds. I also checked the source code, and even tried the condition that checks for the bounds. All seems to be in order. What makes it even more puzzling is that it works perfectly fine for the Ackley function! Am I not seeing something very obvious?

My wrapper is as follows:

import itertools
import numpy as np
import pyswarms as ps
from pyswarms.utils.functions import single_obj as fx

def run_global_best_pso(n_dims, test_func, n_inds, n_gens,\
                        initial_positions=None,\
                        c1=0.5, c2=0.3, w=0.9,\
                        random_seed=12345
        ):
        options = {'c1':c1, 'c2':c2, 'w':w}

        optimizer = ps.single.GlobalBestPSO(n_particles=n_inds, dimensions=n_dims, options=options)
        if initial_positions is not None:
                optimizer.pos = np.array(initial_positions).copy()

        stats = optimizer.optimize(test_func, iters=n_gens)
        pos_history = optimizer.get_pos_history

        return(pos_history)

if __name__ == "__main__":
        n_inds = 100
        n_gens = 10000

        initial_positions = list(itertools.repeat([2.0, 3.0, -3.0], n_inds))

        run_global_best_pso(n_dims=3, test_func=fx.rastrigin_func,\
                                n_inds=n_inds, n_gens=n_gens,\
                                initial_positions=initial_positions)

Update: It works when using different values for the initial positions to solve Rastrigin, e.g., [2.0, 0, -0.02] works but [2.0, 3.0, -0.02] does not.

[ljvmiranda921]

Okay, that's weird. Hmmm, let me check on this later. Thanks for reporting this @szhan !
You mentioned that you want to help in development. Currently, the pipeline is on rewriting the backend of PySwarms for better control. You can check this on PR #115 and #117 . Unfortunately I am still writing my thesis that's why it's a bit stale. Not yet sure which part you can dive-in, currently I'm impelemnting the Ring Topology, maybe you can help by implementing other topologies?

[szhan]

Thanks, @ljvmiranda921 ! I am actually quite interested in topology selection. So, I think I will help with implementing additional topologies first.

[ljvmiranda921]

Awesome, let me update you once I need help. Just need to clean up what I'm doing first then I'll ping you right away!