notev = [0, 1, 3, 4]
currd = {'D': random.choice(bchoice),
'I': random.choice(bchoice),
'S': 0,
'G': random.choice(tchoice),
'L': random.choice(bchoice)}
iterations = int(n/4)
samplesgood = 0
samples = 0
for i in range(0, iterations):
for ne in notev:
# get markov blanket
mb = model.get_markov_blanket(order[ne])
# get evidence of markov balnker
evid = dict((var, currd[var]) for var in mb)
# query
que = infer.query([order[ne]], evidence=evid, show_progress=False)
# random number
ran = random.random()
res = 0
# sort query result
tmp = [(elem, ind) for ind, elem in enumerate(que.values)]
tmp.sort(reverse=True)
# understand new value for variabile
acc = 0
for elem in tmp:
acc += elem[0]
if ran < acc:
def test_markov_blanet(self):
G = BayesianModel([('x', 'y'), ('z', 'y'), ('y', 'w'), ('y', 'v'), ('u', 'w'),
('s', 'v'), ('w', 't'), ('w', 'm'), ('v', 'n'), ('v', 'q')])
self.assertEqual(set(G.get_markov_blanket('y')), set(['s', 'w', 'x', 'u', 'z', 'v']))
