AntColonyTSP/random_search.py at master · YoungseokCh/AntColonyTSP · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
from tsp_parser import Problem
from ant_colony import ACO
import random
import argparse
import sys

parser = argparse.ArgumentParser(
      description = "Hyperparameter Optimisation through Random Search")

parser.add_argument (
      "tsp_file"
    , help    = "Path to .tsp file."
    )

if __name__ == "__main__":
    call_args = parser.parse_args()
    problem = Problem.from_file(call_args.tsp_file)
    best_cost = sys.maxsize

    while True:
        pop_size = random.randrange(problem.dim * 10)+1
        alpha = 1
        beta = 2
        rate = random.uniform(0, 0.8)
        new_hyps = (pop_size, alpha, beta, rate)
        output = ACO(problem, *new_hyps, problem.dim*300)

        print(f"Challenger: {new_hyps}, {output.cost, output.fitness_count}")

        if (best_cost > output.cost):
            best = new_hyps
            best_cost = output.cost
            best_fitness_count = output.fitness_count
        elif (best_cost == output.cost):
            if (best_fitness_count > output.fitness_count):
                best = new_hyps
                best_fitness_count = output.fitness_count

        print(f"{best} are best with cost {best_cost} and fitness count {best_fitness_count}")