def test_tsp_classic():
num_cities = 10
num_iterations = 10
cities = tsp.random_cities(num_cities)
adjacency_matrix = tsp.generate_graph(cities)
solution, penalty = solve_tsp_classic(adjacency_matrix, num_cities,
num_iterations)
assert sorted(solution) == [0] + list(range(num_cities))
assert penalty <= num_cities * sqrt(2)
def test_tsp_classic():
num_cities = 10
num_iterations = 10
cities = tsp.random_cities(num_cities)
adjacency_matrix = tsp.generate_graph(cities)
solution, penalty = solve_tsp_classic(adjacency_matrix, num_cities,
num_iterations)
assert sorted(solution) == [0] + list(range(num_cities))
assert penalty <= num_cities * sqrt(2)
def main_tsp(num_cities=100,
num_iterations=100,
use_islands=False,
use_multiprocess=False,
verbose=True):
'''
This functions creates a random travelling salesman problem and
solves it using classical or island model genetic algorithm.
'''
cities = tsp.random_cities(num_cities)
adjacency_matrix = tsp.generate_graph(cities)
if use_islands:
solution, penalty = solve_tsp_islands(
adjacency_matrix,
num_iterations,
use_multiprocess=use_multiprocess,
verbose=verbose)
else:
solution, penalty = solve_tsp_classic(adjacency_matrix)
print("solution = {0}".format(solution))
print("penalty = {0}".format(penalty))
def create_random_twoopt():
adjacency_matrix = generate_graph(random_cities(10))
two_opt = get_twoopt_mutation(adjacency_matrix, 1.0)
return two_opt, adjacency_matrix
def __init__(self):
cities = [(0.0, 0.0), (1.0, 0.0)]
self.adjacency_matrix = generate_graph(cities)