def test_get_neighbor_simpler_complexity_none0(self):
'''
Possible simpler neighors for 1,2:
1, 0; 1, 2; 0, 2
Run 100 times to reduce probability of randomly passing every time.
'''
for _ in range(1000):
num_models = 3
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node0 = Node(models[1], models[0])
graph.mark_node_visited(node0)
node1 = Node(models[1], models[2])
graph.mark_node_visited(node1)
node = Node(models[0], models[2])
graph.mark_node_visited(node)
neighbor_node_simpler_complexity = graph.get_neighbor_simpler_complexity(node)
self.assertIsNone(neighbor_node_simpler_complexity)
def test_get_neighbor_greater_complexity_none_most_complex_0(self):
'''
Possible more complex neighors for 1,2:
0, 1
1, 0
0, 2
2, 0
Run 1000 times to reduce probability of randomly passing test every time.
'''
for _ in range(1000):
num_models = 3
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[1], models[2])
graph.mark_node_visited(node)
neighbor_node_greater_complexity = graph.get_neighbor_greater_complexity(node)
self.assertIsNone(neighbor_node_greater_complexity)
def test_get_neighbor_simpler_complexity_none(self):
'''
n = 4
Possible simpler neighors for 2, 3:
0, 3
1, 3
2, 1
2, 0
Run 1000 times to reduce probability of randomly passing test every time.
'''
for _ in range(1000):
num_models = 4
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[3], models[2])
graph.mark_node_visited(node)
neighbor_node_simpler = graph.get_neighbor_simpler_complexity(node)
# print(neighbor_node_simpler.complexities)
# valid_set = set([(0, 3), (3, 0), (1, 3), (3, 1)])
valid_set = set([(0, 3), (3, 0), (1, 3), (3, 1), (1, 2), (2, 1), (2, 0), (0, 2)])
self.assertTrue(neighbor_node_simpler.complexities in valid_set)
def test_get_random_neighbor_basic1(self):
'''
Possible random neighors for 3,2:
0,2 3,0
1,2 3,1
4,2 3,4
Run 1000 times to reduce probability of randomly passing every time.
'''
for _ in range(1000):
num_models = 5
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[3], models[2])
graph.mark_node_visited(node)
random_neighbor_node = graph.get_random_neighbor_pairing(node)
valid_set = set([(0, 2), (1, 2), (4, 2), (3, 0), (3, 1), (3, 4)])
self.assertTrue(random_neighbor_node.complexities in valid_set)
def test_graph_init_assertion_error(self):
num_models = 1
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
try:
graph = Graph(models)
except AssertionError:
pass
def test_get_most_simple_node_empty(self):
num_models = 1
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
try:
graph = Graph(models)
except AssertionError:
pass
def test_has_unvisited_simpler_complexity_neighbors_none0(self):
num_models = 3
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[0], models[1])
graph.mark_node_visited(node)
self.assertFalse(graph.has_unvisited_simpler_complexity_neighbors(node))
def test_has_unvisited_greater_complexity_neighbors0(self):
num_models = 5
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[2], models[4])
graph.mark_node_visited(node)
self.assertTrue(graph.has_unvisited_greater_complexity_neighbors(node))
def test_get_most_simple_node_basic1(self):
num_models = 3
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = graph.get_most_simple_node()
is_most_simple = (node.complexities[0] == 0 and node.complexities[1] == 1)\
or (node.complexities[0] == 1 and node.complexities[1] == 0)
self.assertTrue(is_most_simple)
def test_graph_init_basic(self):
num_models = 4
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
# Check to make sure that visited only contains nodes with
# same model complexities
self.assertEqual(num_models, len(graph.visited))
for i in range(num_models):
self.assertTrue((i, i) in graph.visited)
def test_mark_node_visited_basic1(self):
num_models = 12
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[4], models[8])
graph.mark_node_visited(node)
self.assertTrue((8, 4) in graph.visited)
self.assertTrue((4, 8) in graph.visited)
def test_mark_node_visited_basic0(self):
num_models = 4
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[0], models[1])
graph.mark_node_visited(node)
self.assertTrue((0, 1) in graph.visited)
self.assertTrue((1, 0) in graph.visited)
def test_has_unvisited_greater_complexity_neighbors_none(self):
'''
n = 5
3, 4 is most complex node, should have any more complex neighbors
'''
num_models = 5
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[4], models[3])
graph.mark_node_visited(node)
self.assertFalse(graph.has_unvisited_greater_complexity_neighbors(node))
def test_get_neighbor_simpler_complexity_none_most_simple(self):
'''
n = 4
Possible simpler neighors for 0, 1: None
Run 1000 times to reduce probability of randomly passing test every time.
'''
for _ in range(1000):
num_models = 4
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[0], models[1])
graph.mark_node_visited(node)
neighbor_node_simpler = graph.get_neighbor_simpler_complexity(node)
self.assertIsNone(neighbor_node_simpler)
def test_get_neighbor_simpler_complexity_basic0(self):
'''
Possible simpler neighors for 1,2:
1, 0; 1, 2; 0, 2
Run 100 times to reduce probability of randomly passing every time.
'''
for _ in range(1000):
num_models = 3
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[1], models[2])
graph.mark_node_visited(node)
neighbor_node_simpler_complexity = graph.get_neighbor_simpler_complexity(node)
valid_set = set([(1, 0), (2, 1), (0, 2), (2, 0), (1, 2), (0, 1)])
self.assertTrue(neighbor_node_simpler_complexity.complexities in valid_set)
def test_get_random_neighbor_none(self):
'''
Possible random neighors for 3,2:
0,2 3,0
1,2 3,1
4,2 3,4
Run 1000 times to reduce probability of randomly passing every time.
'''
for _ in range(1000):
num_models = 2
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[1], models[0])
graph.mark_node_visited(node)
random_neighbor_node = graph.get_random_neighbor_pairing(node)
self.assertTrue(random_neighbor_node == None)
def test_get_neighbor_greater_complexity_basic1(self):
'''
n = 6
Possible more complex neighors for 4, 3:
5, 3 and 4, 5
Run 100 times to reduce probability of randomly passing every time.
'''
for _ in range(100):
num_models = 6
models = []
for i in range(num_models):
models.append(Model(None))
models[i].complexity_index = i
graph = Graph(models)
node = Node(models[4], models[3])
graph.mark_node_visited(node)
neighbor_node_greater_complexity = graph.get_neighbor_greater_complexity(node)
valid_set = set([(5, 3), (4, 5), (3, 5), (5, 4)])
self.assertTrue(neighbor_node_greater_complexity.complexities in valid_set)
