main.py

from src.algos import SimulatedAnnealing, NearestNeighbour, TabuSearch, HillClimbing
from src.utils import load_data, PathPlotter, BenchmarkPlotter, DistanceHistoryPlotter
from src.algos import MultistartAlgorithm
import os

if __name__ == "__main__":

    distances = load_data(os.path.join("data", "TSP_29.xlsx"))
    NUM_STARTS = 5

    algorithms = [SimulatedAnnealing, NearestNeighbour, TabuSearch, HillClimbing]

    results = [
        min(NearestNeighbour().solve(distances, start_order=x + 1) for x in range(29))
    ]
    results.append(
        MultistartAlgorithm()(
            SimulatedAnnealing,
            distances,
            n_starts=NUM_STARTS,
            temp=1000,
            alpha=0.9,
            n_iter=100,
            verbose=True,
        )
    )
    results.append(
        MultistartAlgorithm()(
            TabuSearch,
            distances,
            n_starts=NUM_STARTS,
            verbose=True,
            tabu_length=3,
            n_iter=30,
        )
    )
    results.append(
        MultistartAlgorithm()(
            HillClimbing, distances, n_starts=NUM_STARTS, verbose=True, n_iter=25
        )
    )

    # Example of plotting the path
    pp = PathPlotter()
    pp.plot(results[0])

    # Example of plotting the benchmark
    bp = BenchmarkPlotter()
    bp.plot(results)

    # Example of plotting the benchmark with custom labels and palette
    bp.plot(
        results, labels=["RAZ", "DWA", "TRZY", "CZTERY"], palette="Purples_d"
    )  # Reds_d, Blues_d, Greens_d, Purples_d, Oranges_d, Greys_d

    dhp = DistanceHistoryPlotter()
    dhp.plot(
        results[1:]
    )  # results[1:] to exclude NearestNeighbour - it has no distance history