def create_qmr(nD, nF, density):
"""
create qmr BN using
nD = number of disease
nF = number of findings
density = proba for disease i to be linked to finding j
"""
name = "qmr%d-%d-%f" % (nD, nF, density)
bn = gum.BayesNet(name)
lDis = [
bn.add(gum.LabelizedVariable("D%4d" % i, "D%4d" % i, 2))
for i in range(nD)
]
# add noisyOR nodes with random leaks
lFind = [
bn.addNoisyOR(gum.LabelizedVariable("F%d" % i, "F%d" % i, 2),
random.uniform(0.1, 0.9)) for i in range(nF)
]
for node in lDis:
prior = random.uniform(0.01, 0.99)
bn.cpt(node)[:] = [prior, 1 - prior]
for i in lDis:
for j in lFind:
if random.random() < density:
# add new cause i to noisyOR j with random weight
bn.addWeightedArc(i, j, random.uniform(0.1, 0.9))
return bn
def createRandomMNTree(nodeNumber, minDomainSize=2, maxDomainSize=2):
"""
Build a random markov network and return it, the returned markov network is a tree
Examples
--------
>>> import markovNetworkLearning as mnl
>>> mn=mnl.createRandomMNTree(15,2,4)
Parameters
----------
nodeNumber : int
the number of variables of the markov network
minDomainSize : int
2 or above, default value : 2, each variable will have a random domain size randomly selected between minDomainSize and maxDomainSize
maxDomainSize : int
2 or above, and more than minDomainSize, default value : 2, each variable will have a random domain size randomly selected between minDomainSize and maxDomainSize
Returns
-------
pyAgrum.MarkovNet
the resulting Markov network
"""
variablesDomainSize = np.random.randint(minDomainSize,
maxDomainSize + 1,
size=nodeNumber)
tree = gum.MarkovNet()
if nodeNumber > 0:
tree.add(gum.LabelizedVariable("0", "", int(variablesDomainSize[0])))
for i in range(1, nodeNumber):
otherNode = random.choice(list(tree.nodes()))
tree.add(gum.LabelizedVariable(str(i), "",
int(variablesDomainSize[i])))
tree.addFactor({otherNode, str(i)})
return tree
def testKL(self):
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 2)
p = gum.Potential().add(v).fillWith([0.0, 1.0])
q = gum.Potential().add(v).fillWith([0.5, 0.5])
r = gum.Potential().add(v).fillWith([0.7, 0.3])
s = gum.Potential().add(v).add(w).fillWith([0.0, 1.0, 0.2, .08])
self.assertEqual(p.KL(p), 0.0)
with self.assertRaises(gum.InvalidArgument):
res = p.KL(s)
with self.assertRaises(gum.InvalidArgument):
res = s.KL(p)
res = p.KL(q)
self.assertAlmostEqual(res, 0.0 + 1.0 * math.log(1.0 / 0.5, 2))
with self.assertRaises(gum.FatalError):
res = q.KL(p)
res = p.KL(r)
self.assertAlmostEqual(res, 0.0 + 1.0 * math.log(1.0 / 0.3, 2))
with self.assertRaises(gum.FatalError):
res = r.KL(p)
self.assertAlmostEqual(
q.KL(r), 0.5 * math.log(0.5 / 0.7, 2) + 0.5 * math.log(0.5 / 0.3, 2))
self.assertAlmostEqual(
r.KL(q), 0.7 * math.log(0.7 / 0.5, 2) + 0.3 * math.log(0.3 / 0.5, 2))
def testLogit(self):
bn = gum.BayesNet()
age = bn.add(gum.RangeVariable("age", "", 35, 67))
taux = bn.add(gum.RangeVariable("taux", "", 115, 171))
angine = bn.add(gum.LabelizedVariable("angine", ""))
vc = gum.LabelizedVariable("coeur", "", 0)
vc.addLabel("NON").addLabel("OUI")
coeur = bn.addLogit(vc, 14.4937)
bn.addWeightedArc(age, coeur, -0.1256)
bn.addWeightedArc(taux, coeur, -0.0636)
bn.addWeightedArc(angine, coeur, 1.779)
witness_age = ("50", "49", "46", "49", "62", "35", "67", "65", "47")
witness_taux = ("126", "126", "144", "139", "154", "156", "160", "140",
"143")
witness_angine = ("1", "0", "0", "0", "1", "1", "0", "0", "0")
witness_coeur = ("OUI", "OUI", "OUI", "OUI", "OUI", "OUI", "NON",
"NON", "NON")
witness_proba = (0.8786, 0.5807, 0.3912, 0.3773, 0.2127, 0.8760,
1 - 0.0163, 1 - 0.0710, 1 - 0.3765)
inst = gum.Instantiation(bn.cpt(coeur))
for i in range(len(witness_age)):
inst.chgVal(bn.variable(age), bn.variable(age)[witness_age[i]])
inst.chgVal(bn.variable(taux), bn.variable(taux)[witness_taux[i]])
inst.chgVal(bn.variable(angine),
bn.variable(angine)[witness_angine[i]])
inst.chgVal(bn.variable(coeur),
bn.variable(coeur)[witness_coeur[i]])
self.assertAlmostEqual(bn.cpt(coeur).get(inst),
witness_proba[i],
places=3)
def testVariableInsertion(self):
pot = gum.Potential()
self.assertTrue(pot.empty())
pot.add(self.var['c'])
self.assertFalse(pot.empty())
self.assertEqual(pot.nbrDim(), 1)
pot.add(self.var['s']).add(self.var['r'])
self.assertEqual(pot.nbrDim(), 3)
for id, var in enumerate([self.var['c'], self.var['s'],
self.var['r']]):
self.assertTrue(pot.contains(var))
self.assertEqual(pot.variable(id), var)
self.assertFalse(pot.contains(gum.LabelizedVariable("a", "", 5)))
a = gum.LabelizedVariable("a", "a", 2)
other_a = gum.LabelizedVariable("a", "a", 2)
p = gum.Potential()
p.add(a)
with self.assertRaises(gum.DuplicateElement):
p.add(a) # once again
with self.assertRaises(gum.DuplicateElement):
p.add(other_a) # with the same name
def testLabelsOfVars(self):
v = gum.LabelizedVariable("a", "a")
self.assertEqual(v.labels(), ("0", "1"))
self.assertNotEqual(v.labels(), ("1", "0"))
v = gum.LabelizedVariable(
"b", "b", 0).addLabel("toto").addLabel("titi").addLabel("yes")
self.assertEqual(v.labels(), ("toto", "titi", "yes"))
v = gum.RangeVariable("c", "c", 0, 5)
self.assertEqual(v.labels(), ("0", "1", "2", "3", "4", "5"))
v = gum.RangeVariable("d", "d", 3, 5)
self.assertEqual(v.labels(), ("3", "4", "5"))
v = gum.DiscretizedVariable(
"e", "e").addTick(1).addTick(2).addTick(3).addTick(4)
self.assertEqual(v.labels(), ("[1;2[", "[2;3[", "[3;4]"))
v = gum.DiscretizedVariable(
"e", "e").addTick(1).addTick(2).addTick(3).addTick(4)
v.setEmpirical(True)
self.assertEqual(v.labels(), ("(1;2[", "[2;3[", "[3;4)"))
v = gum.DiscretizedVariable("f", "f", [1, 5, 2, 4])
self.assertEqual(v.labels(), ("[1;2[", "[2;4[", "[4;5]"))
v = gum.DiscretizedVariable("f", "f", [1, 5, 2, 4])
v.setEmpirical(True)
self.assertEqual(v.labels(), ("(1;2[", "[2;4[", "[4;5)"))
def restructure_bn_fenton(self, bn): # could be streamlined
bn.erase('constraint')
list_of_scn_node_names = ["NA"]
exhaustivity_check = 0
for i in bn.nodes():
if 'scn' in bn.variable(i).name():
list_of_scn_node_names.append(bn.variable(i).name())
exhaustivity_check = exhaustivity_check + bn.cpt(i)[0]
if exhaustivity_check != 1:
# add a leak state
scn_3 = bn.add(gum.LabelizedVariable('scn_leak', 'leak', self.truth_values))
bn.cpt(scn_3) [{'scn_leak': self.truth_values[0]}] = 1 - exhaustivity_check
bn.cpt(scn_3)[{'scn_leak': self.truth_values[1]}] = exhaustivity_check
list_of_scn_node_names.append(bn.variable('scn_leak').name())
n_states_in_constraint_node = len(list_of_scn_node_names) # this is n(scn) + 1
aux = bn.add(gum.LabelizedVariable('aux', 'aux', list_of_scn_node_names))
constraint = bn.add(
gum.LabelizedVariable('constraint', 'constraint', self.truth_values)) # aux always has two states
# all nodes are added now... Now add links between the nodes.
list_of_scn_node_names.remove("NA")
self.scenario_nodes = list_of_scn_node_names
for node_name in list_of_scn_node_names:
bn.addArc(node_name, 'aux')
bn.addArc('aux', 'constraint')
self.implement_aux(bn)
self.implement_constraint(bn)
def setUp(self):
self.bn = gum.BayesNet()
self.buffer0 = ""
self.buffer1 = ""
self.var1 = gum.LabelizedVariable("a", "le noeud a", 2)
self.var2 = gum.LabelizedVariable("b", "le noeud b", 2)
self.var3 = gum.LabelizedVariable("c", "le noeud c", 3)
def testEqualities(self):
u = gum.LabelizedVariable("u", "u", 4)
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 3)
p = gum.Potential().add(u).add(v).add(w)
p.random()
q = gum.Potential(p)
self.assertEqual(p, q)
i = gum.Instantiation(q)
i.setLast()
q[i] = 0
self.assertNotEqual(p, q)
q.fillWith(p)
self.assertEqual(p, q)
q.fillWith(1)
self.assertNotEqual(p, q)
q.fillWith(p)
self.assertEqual(p, q)
x = gum.LabelizedVariable("x", "Unknown", 5)
q.add(x)
self.assertNotEqual(p, q)
q = gum.Potential().add(v).add(w)
self.assertNotEqual(p, q)
def setUp(self):
self.var1 = gum.LabelizedVariable("var1", "1", 2)
self.var2 = gum.LabelizedVariable("var2", "2", 2)
self.var3 = gum.LabelizedVariable("var3", "3", 2)
self.var4 = gum.LabelizedVariable("var4", "4", 2)
self.var5 = gum.LabelizedVariable("var5", "(gum::Size) 5", 3)
self.bufferlisten = ""
self.bufferecoute = ""
def testZeroParent(self):
bn = gum.BayesNet()
a = bn.addOR(gum.LabelizedVariable('a', 'a', 2))
self.assertEqual(bn.cpt(a).tolist(), [1.0, 0.0])
b = bn.addAND(gum.LabelizedVariable('b', 'b', 2))
self.assertEqual(bn.cpt(b).tolist(), [0.0, 1.0])
def setUp(self):
self.var = {}
self.var.update({"c": gum.LabelizedVariable("c", "nuages", 2)})
self.var.update({"s": gum.LabelizedVariable("s", "arrosoir", 2)})
r = gum.LabelizedVariable("r", "pluie", 0).addLabel('bof') \
.addLabel('carrement')
self.var.update({"r": r})
self.var.update({"w": gum.LabelizedVariable("w", "herbe mouillée", 2)})
def testOperationWithDifferentVariablesFromMadsLindskou(self):
px = gum.Potential()
py = gum.Potential()
for s in "ab":
px.add(gum.LabelizedVariable(s, s, 2))
for s in "bc":
py.add(gum.LabelizedVariable(s, s, 2))
def testHashableDiscreteVariable(self):
va = gum.LabelizedVariable("a", "a")
vb = gum.LabelizedVariable("b", "b", ["toto", "titi", "yes"])
vc = gum.RangeVariable("c", "c", 0, 5)
vd = gum.RangeVariable("d", "d", 3, 5)
ve = gum.DiscretizedVariable(
"e", "e").addTick(1).addTick(2).addTick(3).addTick(4)
s = set([va, vb, vc] + [vc, vd, ve] + [va, ve])
self.assertEqual(len(s), 5)
def testRandomPotential(self):
u = gum.LabelizedVariable("u", "u", 4)
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 3)
p = gum.Potential().add(u).add(v).add(w)
I = gum.Instantiation(p)
p.random()
I.setFirst()
while not I.end():
self.assertTrue(p.get(I) <= 1.0)
I.inc()
p.randomDistribution()
I.setFirst()
cum = 0
while not I.end():
self.assertTrue(p.get(I) <= 1.0)
cum += p.get(I)
I.inc()
self.assertAlmostEqual(cum, 1.0, 6)
p.randomCPT()
v = p.variable(0)
I.setFirst()
while not I.end():
cum = 0.0
I.setFirstVar(v)
while not I.end():
self.assertTrue(p.get(I) <= 1.0)
cum += p.get(I)
I.incVar(v)
self.assertAlmostEqual(cum, 1.0, 6)
I.unsetEnd()
I.incNotVar(v)
p.fillWith(1).normalizeAsCPT()
I.setFirst()
while not I.end():
self.assertAlmostEqual(p.get(I), 0.25, 6)
I.inc()
alpha = 0.0
while alpha <= 1.0:
p.fillWith(1).normalizeAsCPT()
p.noising(alpha)
min = (1 - alpha) * 0.25 + alpha * 0.0
max = (1 - alpha) * 0.25 + alpha * 1.0
I.setFirst()
while not I.end():
self.assertTrue(min <= p.get(I) <= max)
I.inc()
alpha += 0.1
def testVariableAccessor(self):
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 2)
p = gum.Potential().add(v).add(w)
self.assertEqual(p.variable(0), p.variable('v'))
self.assertEqual(p.variable(1), p.variable('w'))
self.assertNotEqual(p.variable(1), p.variable('v'))
self.assertNotEqual(p.variable(0), p.variable('w'))
with self.assertRaises(gum.NotFound):
x = p.variable("zz")
def testInverse(self):
u = gum.LabelizedVariable("u", "u", 4)
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 3)
p = gum.Potential().add(u).add(v).add(w)
p.random()
q = gum.Potential(p).inverse()
self.assertAlmostEqual((q * p).max(), 1.0, delta=1e-7)
self.assertAlmostEqual((q * p).min(), 1.0, delta=1e-7)
def testArgMaxMinFindAll(self):
v = gum.LabelizedVariable("v", "v", 2)
w = gum.LabelizedVariable("w", "w", 3)
p = gum.Potential()
self.__test_val_for_set(p, 10, p.findAll(10), 0)
p.add(v).add(w)
p.fillWith([1, 3, 2, 4, 1, 4])
self.__test_val_for_set(p, 3, p.findAll(3), 1)
self.__test_val_for_set(p, 10, p.findAll(10), 0)
self.__test_val_for_set(p, 4, p.argmax()[0], 2)
self.__test_val_for_set(p, 1, p.argmin()[0], 2)
def testEntropyPotential(self):
a = gum.LabelizedVariable("a", "a", 2)
p = gum.Potential().add(a)
self.assertEqual(p.fillWith([0, 1]).entropy(), 0.0)
self.assertEqual(p.fillWith([1, 0]).entropy(), 0.0)
self.assertEqual(p.fillWith([0.5, 0.5]).entropy(), 1.0)
def testReorganizePotential(self):
a, b, c, d = [gum.LabelizedVariable(s, s, 3) for s in "abcd"]
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
q = gum.Potential()
q.add(c).add(d)
q.fillWith([4, 5, 6, 3, 2, 1, 4, 3, 2])
self.assertNotEqual(str(p * q), str(q * p))
self.assertEqual(str(p * q), str((q * p).reorganize(['c', 'd', 'a', 'b'])))
self.assertNotEqual(
str(p * q), str((q * p).reorganize(['c', 'a', 'd', 'b'])))
try:
q.reorganize(['a'])
self.assertTrue(False)
except gum.InvalidArgument:
self.assertTrue(True)
try:
q.reorganize(['d'])
self.assertTrue(False)
except gum.InvalidArgument:
self.assertTrue(True)
def testReorganizePotentialWithAccents(self):
if sys.version_info >= (3, 0):
a, b, c, d = [gum.LabelizedVariable(s, s, 3) for s in "éèàû"]
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
q = gum.Potential()
q.add(c).add(d)
q.fillWith([4, 5, 6, 3, 2, 1, 4, 3, 2])
self.assertNotEqual(str(p * q), str(q * p))
self.assertEqual(
str(p * q), str((q * p).reorganize(['à', 'û', 'é', 'è'])))
self.assertNotEqual(
str(p * q), str((q * p).reorganize(['à', 'é', 'û', 'è'])))
try:
q.reorganize(['é'])
self.assertTrue(False)
except gum.InvalidArgument:
self.assertTrue(True)
try:
q.reorganize(['û'])
self.assertTrue(False)
except gum.InvalidArgument:
self.assertTrue(True)
def read_file(filename):
"""
Renvoie les variables aléatoires et la probabilité contenues dans le
fichier dont le nom est passé en argument.
"""
Pjointe = gum.Potential()
variables = []
fic = open(filename, 'r')
# on rajoute les variables dans le potentiel
nb_vars = int(fic.readline())
for i in range(nb_vars):
name, domsize = fic.readline().split()
variable = gum.LabelizedVariable(name, name, int(domsize))
variables.append(variable)
Pjointe.add(variable)
# on rajoute les valeurs de proba dans le potentiel
cpt = []
for line in fic:
cpt.append(float(line))
Pjointe.fillWith(np.array(cpt))
fic.close()
return np.array(variables), Pjointe
def testEquality(self):
a, b, c = [gum.LabelizedVariable(s, s, 3) for s in "abc"]
q = gum.Potential()
q.add(b).add(c)
q.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
p = gum.Potential() # same data, difference dims
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
r = gum.Potential() # same data, same dims
r.add(a).add(b)
r.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
t = gum.Potential() # same dims, different data
t.add(a).add(b)
t.fillWith([1, 2, 3, 0, 5, 6, 7, 8, 9])
u = gum.Potential() # same dims, same data, different order
u.add(b).add(a)
u.fillWith([1, 4, 7, 2, 5, 8, 3, 6, 9])
self.assertTrue(p == p)
self.assertFalse(p == q)
self.assertTrue(p == r)
self.assertFalse(p == t)
self.assertTrue(p == u)
self.assertFalse(p != p)
self.assertTrue(p != q)
self.assertFalse(p != r)
self.assertTrue(p != t)
self.assertFalse(p != u)
def undiGraphToMarkovNetwork(graph, variables=dict(), domainSize=2):
"""
Build a markov network from an UndiGraph
Variables can be either given in parameters or can be created if they are not defined
Parameters
----------
graph : pyAgrum.UndiGraph
the graph
variables : dict<str:pyAgrum.DiscreteVariable>
already defined variables, keys are variables name
domainSize : int
the domain of variables that are not defined
Returns
-------
pyAgrum.MarkovNet
the resulting Markov network
"""
nodes = graph.nodes()
cliques = getAllMaximalCliquesUndiGraph(graph)
definedVariables = variables.keys()
mn = gum.MarkovNet()
for node in nodes:
if node in definedVariables:
mn.add(variables[node])
else:
mn.add(gum.LabelizedVariable(str(node), "", domainSize))
for clique in cliques:
mn.addFactor(clique)
return mn
def testMargOutOperators(self):
a, b, c, d = [gum.LabelizedVariable(s, s, 3) for s in "abcd"]
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
p.normalize()
q = gum.Potential()
q.add(c).add(d)
q.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
q.normalize()
joint = p * q
margAB = joint.margSumOut(["c", "d"])
self.assertEqual(margAB.names, p.names)
self.assertEqual(margAB.tolist(), p.tolist())
margCD = joint.margSumOut(
["b", "a", "x"]) # note the vars in a different order and with one not present in the potential
self.assertEqual(margCD.names, q.names)
self.assertEqual(margCD.tolist(), q.tolist())
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
self.assertEqual(p.margProdOut(["a"]).tolist(), [6, 120, 504])
self.assertEqual(p.margProdOut(["b"]).tolist(), [28, 80, 162])
self.assertEqual(p.margMaxOut(["a"]).tolist(), [3, 6, 9])
self.assertEqual(p.margMaxOut(["b"]).tolist(), [7, 8, 9])
self.assertEqual(p.margMinOut(["a"]).tolist(), [1, 4, 7])
self.assertEqual(p.margMinOut(["b"]).tolist(), [1, 2, 3])
def testDimensionIncreasing(self):
bn = gum.BayesNet()
id_list = []
self.fillBN(bn, id_list)
bn.cpt(id_list[3]).fillWith([1, 0,
0.1, 0.9,
0.1, 0.9,
0.01, 0.99])
addvar = gum.LabelizedVariable("d", "rosée", 4)
addvar_id = bn.add(addvar)
bn.addArc(addvar_id, id_list[3])
list3 = bn.cpt(id_list[3])
self.assertEqual(list3.names, ('w', 's', 'r', 'd'))
list3[0, :, :, :] = \
[[[0.99, 0.1], [0.1, 0]], [[0.01, 0.9], [0.9, 1]]]
list3[1, :, :, :] = \
[[[0.99, 0.1], [0.1, 0]], [[0.01, 0.9], [0.9, 1]]]
list3[2, :, :, :] = \
[[[0.99, 0.1], [0.1, 0]], [[0.01, 0.9], [0.9, 1]]]
list3[3, :, :, :] = \
[[[0.99, 0.1], [0.1, 0]], [[0.01, 0.9], [0.9, 1]]]
list3[2, :, :, :] = 0.25
self.assertListsAlmostEqual(list3[2, :, :, :].tolist(),
[[[0.25, 0.25], [0.25, 0.25]], [[0.25, 0.25], [0.25, 0.25]]])
self.assertListsAlmostEqual(list3[{"d": 1}].tolist(),
[[[0.99, 0.1], [0.1, 0]], [[0.01, 0.9], [0.9, 1]]])
def testSimpleInPLaceOperators(self):
a, b, c = [gum.LabelizedVariable(s, s, 3) for s in "abc"]
q = gum.Potential()
q.add(b).add(c)
q.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
z = p + q
p += q
self.assertEqual(z.tolist(), p.tolist())
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
z = p - q
p -= q
self.assertEqual(z.tolist(), p.tolist())
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
z = p * q
p *= q
self.assertEqual(z.tolist(), p.tolist())
p = gum.Potential()
p.add(a).add(b)
p.fillWith([1, 2, 3, 4, 5, 6, 7, 8, 9])
z = p / q
p /= q
self.assertEqual(z.tolist(), p.tolist())
def dag_to_bn(dag, names):
# DAG to BN
bn = gum.BayesNet()
for name in names:
bn.add(gum.LabelizedVariable(name))
for arc in dag.arcs():
bn.addArc(arc[0], arc[1])
return bn
def implement_aux_constraint_structure(self, bn):
list_of_scn_node_names = ["NA"]
for i in bn.nodes():
if 'scn' in bn.variable(i).name():
list_of_scn_node_names.append(bn.variable(i).name())
n_states_in_constraint_node = len(list_of_scn_node_names) # this is n(scn) + 1
aux = bn.add(gum.LabelizedVariable('aux', 'aux', list_of_scn_node_names))
constraint = bn.add(
gum.LabelizedVariable('constraint', 'constraint', self.truth_values)) # aux always has two states
# all nodes are added now... Now add links between the nodes.
list_of_scn_node_names.remove("NA")
self.scenario_nodes = list_of_scn_node_names
for node_name in list_of_scn_node_names:
bn.addArc(node_name, 'aux')
bn.addArc('aux', 'constraint')
self.implement_aux(bn)
self.implement_constraint(bn)
def add_node(self, code):
"""Add a new node to the network."""
node = self.twodbn.add(gum.LabelizedVariable(code, code, 2),
self.code_as_number(code))
self.nodes.append(node)
return node
