How to run two optimizations in parallel

I am trying to optimize a function which takes a long time. There are many different global optimizers so I would like to run them in parallel, seeing the progress they make in different windows. Here is my non-parallel code:

##### optimization 1 ####
from scipy.optimize import basinhopping

def build_show_bh(MIN=None):
    if MIN is None:
        MIN = [0]
    def fn(xx, f, accept):
        if f < MIN[-1]:
            print([round(x, 2) for x in xx], f)
            MIN.append(f)
    return fn

x0 = (0.5, 0.5, 0.5)
bounds = [(0,1)]*3
minimizer_kwargs = dict(method="L-BFGS-B", bounds=bounds)
progress_f = [0]
c = build_show_bh(progress_f)
print("Optimizing using basinhopping")
res = basinhopping(
    opt, 
    x0,
    minimizer_kwargs=minimizer_kwargs,
    niter=10,
    callback=c, 
    disp=True
)
print(f"external way of keeping track of MINF: {progress_f}")

##### optimization 2 ####

import dlib
lower_bounds = [0]*3
upper_bounds = [1]*3
x, y = dlib.find_max_global(opt_dlib, lower_bounds, upper_bounds, 10)
print(f"The optimal inputs are {x} with value {y}")

How can I do this?

I am on Python 3.10.

You can try using a ProcessPoolExecutor to lift the GIL and run each optimization as a separate proces:

from threading import Lock

print_lock = Lock()

from scipy.optimize import basinhopping


def scipy_optimize():
    def build_show_bh(MIN=None):
        if MIN is None:
            MIN = [0]
        def fn(xx, f, accept):
            if f < MIN[-1]:
                with print_lock:
                    print([round(x, 2) for x in xx], f)
                MIN.append(f)
        return fn

    x0 = (0.5, 0.5, 0.5)
    bounds = [(0,1)]*3
    minimizer_kwargs = dict(method="L-BFGS-B", bounds=bounds)
    progress_f = [0]
    c = build_show_bh(progress_f)
    with print_lock:
        print("Optimizing using basinhopping")

    res = basinhopping(
        opt, 
        x0,
        minimizer_kwargs=minimizer_kwargs,
        niter=10,
        callback=c, 
        disp=True
    )
    with print_lock:
        print(f"external way of keeping track of MINF: {progress_f}")

##### optimization 2 ####

import dlib

def dlib_optimize():
    lower_bounds = [0]*3
    upper_bounds = [1]*3
    x, y = dlib.find_max_global(opt_dlib, lower_bounds, upper_bounds, 10)
    with print_lock:
        print(f"The optimal inputs are {x} with value {y}")

with concurrent.futures.ProcessPoolExecutor() as executor:
    executor.submit(scipy_optimize)
    executor.submit(dlib_optimize)