main.py

import tabu_search
import obj_function
import load_data
import numpy as np
import timeit
import random
import sys

if __name__ == "__main__":
    tabu = [1, 10, 100, 350, 600, 1000]
    for i in range(4):

        # BEGIN - Parameters for algorithm 
        file_name = 'pcb.txt'
        point_number = 10
        number_of_tools = 3
        t_min = 3
        number_of_iterations = 1000000
        tabu_list_length = tabu[i]
        type_of_neighborhood = 4
        seed = None
        # END - Parameters for algorithm  

        init_solution = np.arange(1, point_number + 1)
        #np.random.shuffle(init_solution)

        OF = obj_function.ObjectiveFun(file_name, t_min, point_number, number_of_tools)

        TS = tabu_search.TabuSearch(file_name, init_solution, number_of_iterations, tabu_list_length,
                                    type_of_neighborhood, t_min, point_number, number_of_tools, seed)

        start = timeit.default_timer()
        TS.run()
        stop = timeit.default_timer()

        print("*****************************************")
        # print("Initial solution: \n" + str(init_solution))
        print("Initial objective function value: " + "{0:.2f}".format(OF.obj_function(init_solution)))
        print("-----------------------------------------")
        # print("Best solution: \n" + str(TS.solution) + "\n")
        # print("Best real solution: \n" + str(OF.real_permutation(TS.solution)) + "\n")
        print("Best objective function value: " + "{0:.2f}".format(TS.objective_fun_value))
        print("Time of execution: " + "{0:.2f}".format(stop - start) + " seconds")
        print("*****************************************\n")


        TS.save_objective_fun_value_plots()

        print("Record successfully saved to records.csv file!")
        print("Figures saved to Fig/ directory!")

        # TS.show_objective_fun_value_plots()