def test_nearest_neighbor_heuristic_case_1(self):
# 1 _ _ 4 _ 3 _ 0 _ _ _ 2 ( "_" = 1 unit of distance) (symmetric case)
distance_matrix_case_1 = np.array([[0, 4, 3, 1, 2],
[4, 0, 7, 3, 2],
[3, 7, 0, 4, 5],
[1, 3, 4, 0, 1],
[2, 2, 5, 1, 0]], dtype=int)
ordered_path_case_1 = OrderedPath(np.array([0, 3, 4, 1, 2], dtype=int), distance_matrix_case_1)
expected_total_weight_case_1 = 14 # d = 1+1+2+7+3 = 14
nearest_neighbor_case_1 = NearestNeighborHeuristic(distance_matrix_case_1)
self.assertEqual(ordered_path_case_1.distance(), expected_total_weight_case_1)
self.assertEqual(nearest_neighbor_case_1.get_ordered_path(), ordered_path_case_1)
self.assertEqual(nearest_neighbor_case_1.get_total_weight(), expected_total_weight_case_1)
def test_two_opt_improvement_heuristic_case_2(self):
# 1 _ _ 4 _ 3 _ 0 _ _ _ 2 ( "_" = 1 unit of distance) (symmetric case) (best result: d=14)
distance_matrix_case_2 = np.array(
[[0, 4, 3, 1, 2], [4, 0, 7, 3, 2], [3, 7, 0, 4, 5],
[1, 3, 4, 0, 1], [2, 2, 5, 1, 0]],
dtype=int)
# d = 4+7+4+1+2 = 17
ordered_path_case_2 = OrderedPath(np.array([0, 1, 2, 3, 4], dtype=int),
distance_matrix_case_2)
case_2 = TwoOptImprovementHeuristic(ordered_path_case_2)
self.assertTrue(
case_2.get_total_weight() < ordered_path_case_2.distance())
ordered_path_case_1 = OrderedPath(np.array([0, 3, 4, 1, 2], dtype=int),
distance_matrix_case_2) # d = 14
self.assertTrue(case_2.get_total_weight() >= 14)