def test_map_medium_bn(self):
medium_bn = parse_network_from_file(
"./sample_bayesian_networks/insurance.bif")
prob, assgn = map_ask(dict(PropCost="Thousand",
RiskAversion="Adventurous"),
medium_bn,
map_vars=['Theft'])
self.assertEqual(prob, 0.13093217869278678)
self.assertEqual(assgn, [('Theft', 'False')])
def test_map_small_bn_2(self):
small_bn = parse_network_from_file(
"./sample_bayesian_networks/earthquake.bif")
prob, assgn = map_ask(dict(JohnCalls="True", Alarm="True"),
small_bn,
not_map_vars=['Alarm'])
self.assertEqual(prob, 0.00580356)
self.assertEqual(assgn, [('Burglary', 'True'), ('Earthquake', 'False'),
('MaryCalls', 'True')])
def plot_map_time_growing_var_fixed_evidence(bn_path):
bn = parse_network_from_file(bn_path)
variables = bn.variables
var_values = [bn.variable_values(var) for var in variables]
random_sel_values = [random.choice(val_list) for val_list in var_values]
dic = dict(zip(variables, random_sel_values))
times_list = []
first_n_pairs = {k: dic[k] for k in list(dic)[:2]}
print(first_n_pairs)
del variables[:2]
print(variables)
for i, var in enumerate(variables):
bn.last_max_out = None
map_vars = variables[:i]
if map_vars:
ts = time.time()
map_ask(first_n_pairs, bn, map_vars=map_vars)
te = time.time()
times_list.append(te - ts)
plot_list(times_list, "MAP growing map vars fixed evidence", "# map vars",
"time")
def plot_map_chain_var_network_size():
times_list = []
for i in range(4, 500):
size = i
bn = generate_chain(size=size)
ts = time.time()
map_vars = [F"S{r}" for r in range(int(size / 2), size - 1)]
prob, assgn = map_ask(dict([(F"S{size}", "True")]),
bn,
not_map_vars=["S0"])
te = time.time()
times_list.append(te - ts)
plot_list(times_list, "MAP on chain bayesian networks",
"size of the chain", "computation time")
def plot_map_insurance_network_var_mapvars():
times_list = []
bn = parse_network_from_file("./sample_bayesian_networks/insurance.bif")
for i in range(1, len(bn.variables)):
size = i
bn = parse_network_from_file(
"./sample_bayesian_networks/insurance.bif")
map_vars = bn.variables[0:i]
ts = time.time()
prob, assgn = map_ask(dict(), bn, map_vars=map_vars)
te = time.time()
times_list.append(te - ts)
plot_list(times_list, "MAP on Insurance network", "# of MAP variables",
"computation time")
def plot_map_var_paper_network_size():
times_list = []
for i in range(4, 15):
size = i
bn = parse_network(generate_polytree_network_bif(size=size))
map_vars = [F"X{r}" for r in range(1, size + 1)]
ts = time.time()
prob, assgn = map_ask(dict([(F"S{size}", "TRUE")]),
bn,
map_vars=map_vars)
te = time.time()
times_list.append(te - ts)
plot_list(times_list, "MAP on dynamic bayesian networks",
"size of the network", "computation time")
def test_map_paper_network():
size = 17
bn = parse_network(generate_polytree_network_bif(size=size))
map_vars = [F"X{r}" for r in range(1, size + 1)]
prob, assgn = map_ask(dict([(F"S{size}", "TRUE")]), bn, map_vars=map_vars)
print(F"P(map,e): {prob}, MAP: {assgn}")
def test_map():
bn = parse_network_from_file("./sample_bayesian_networks/earthquake.bif")
prob, assgn = map_ask(dict(Burglary="True", JohnCalls="True"),
bn,
not_map_vars=['Alarm'])
print(F"P(map,e): {prob}, MAP: {assgn}")