main.py

import algorithms as algo
from utils import tools
from os import listdir
from os.path import isfile, join
from os import path
import argparse
import csv
import time
import matplotlib.pyplot as plt
from pathlib import Path
import numpy as np


best_known = {
    'tai20a': 703482,
    'tai40a': 3139370,
    'tai60a': 7205962,
    'tai80a': 13499184,
    'tai100a': 21044752,
}


def save_cost_history_plot(history, file, method,n_iter):
    plt.figure(figsize=(10, 5))
    x = [0, n_iter]
    y = [best_known[file]] * len(x)
    plt.plot(history, label='cost_function')
    plt.plot(x, y,  'go-', label='best_known')
    min_value = min(history)
    min_idx = history.index(min_value)
    plt.plot(min_idx, min_value, 'ro', label='found_optimum')
    plt.grid()
    plt.legend()
    plt.title(method)
    Path(f"./images").mkdir(parents=True, exist_ok=True)
    plt.savefig(f'./images/{method}_{file}.png')
    print(f'./images/{method}_{file}.png saved!')


def main(args):
    n_iter = 100
    if args.make_csv:
        with open('statistic.csv', 'w') as file:
            columns_names = ['File name', 'Method', 'Best known', 'Result', 'Time']
            writer = csv.DictWriter(file, fieldnames=columns_names)
            writer.writeheader()
            reader = tools.QAReader()
            benchmarks = [f for f in listdir(args.path) if isfile(join(args.path, f))]
            for file in benchmarks:
                data = reader(path.join(args.path, file))
                algorithms = [(name, f(data, 'first-with-delta', True, n_iter)) for name, f in algo.__dict__.items() if callable(f)]
                for name, algorithm in algorithms:
                    Path(f"./{name}").mkdir(parents=True, exist_ok=True)
                    print(f'{name} working on {file}')
                    start_time = time.time()
                    solution, final_cost = algorithm()
                    work_time = round(time.time() - start_time, 4)
                    np.savetxt(f'./{name}/{file}.sol', solution.reshape(1, solution.shape[0]), fmt='%d')
                    if name != 'LocalSearch':
                        print(algorithm.cost_history)
                        save_cost_history_plot(algorithm.cost_history, file, name, n_iter)
                    writer.writerow({
                        'File name': file,
                        'Method': name,
                        'Best known': best_known[file],
                        'Result': final_cost,
                        'Time': work_time
                    })


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument("-p", "--path", type=str,
                        help="path to benchmark data", default='./data')
    parser.add_argument("-csv", "--make_csv", action='store_true',
                        help="path to benchmark data", default=True)
    args = parser.parse_args()
    main(args)