def export_evidence_to_uai(self, path, evidence, format):
f = open(path, "w+")
if format == "pomegranate":
import json
# representation of the BN
model_dict = json.loads(self.model.to_json())
# immutable index of variables
index = {k: v for v, k in enumerate(self.node_names)}
inverse_index = {k: v for k, v in enumerate(self.node_names)}
if format == "pgmpy":
from pgmpy.readwrite import UAIWriter
writer = UAIWriter(self.model_pgmpy)
domain = writer.get_domain()
domain = sorted(domain.items(), key=lambda x: (x[1], x[0]))
# number of variables in the sample
f.write(str(len(evidence)) + "\n")
# state/value pairs
for state in evidence:
if format == "pomegranate":
f.write(" " + str(index[state]) + " " + str(evidence[state]) +
"\n")
if format == "pgmpy":
f.write(" " + str([x[0] for x in domain].index(state)) + " " +
str(evidence[state]) + "\n")
return
def setUp(self):
self.maxDiff = None
edges = [['family-out', 'dog-out'],
['bowel-problem', 'dog-out'],
['family-out', 'light-on'],
['dog-out', 'hear-bark']]
cpds = {'bowel-problem': np.array([[0.01],
[0.99]]),
'dog-out': np.array([[0.99, 0.01, 0.97, 0.03],
[0.9, 0.1, 0.3, 0.7]]),
'family-out': np.array([[0.15],
[0.85]]),
'hear-bark': np.array([[0.7, 0.3],
[0.01, 0.99]]),
'light-on': np.array([[0.6, 0.4],
[0.05, 0.95]])}
states = {'bowel-problem': ['true', 'false'],
'dog-out': ['true', 'false'],
'family-out': ['true', 'false'],
'hear-bark': ['true', 'false'],
'light-on': ['true', 'false']}
parents = {'bowel-problem': [],
'dog-out': ['bowel-problem', 'family-out'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']}
self.bayesmodel = BayesianModel(edges)
tabular_cpds = []
for var, values in cpds.items():
cpd = TabularCPD(var, len(states[var]), values,
evidence=parents[var],
evidence_card=[len(states[evidence_var])
for evidence_var in parents[var]])
tabular_cpds.append(cpd)
self.bayesmodel.add_cpds(*tabular_cpds)
self.bayeswriter = UAIWriter(self.bayesmodel)
edges = {('var_0', 'var_1'), ('var_0', 'var_2'), ('var_1', 'var_2')}
self.markovmodel = MarkovModel(edges)
tables = [(['var_0', 'var_1'],
['4.000', '2.400', '1.000', '0.000']),
(['var_0', 'var_1', 'var_2'],
['2.2500', '3.2500', '3.7500', '0.0000', '0.0000', '10.0000',
'1.8750', '4.0000', '3.3330', '2.0000', '2.0000', '3.4000'])]
domain = {'var_1': '2', 'var_2': '3', 'var_0': '2'}
factors = []
for table in tables:
variables = table[0]
cardinality = [int(domain[var]) for var in variables]
values = list(map(float, table[1]))
factor = DiscreteFactor(variables, cardinality, values)
factors.append(factor)
self.markovmodel.add_factors(*factors)
self.markovwriter = UAIWriter(self.markovmodel)
def daoopt_solver(uai_path, evid_path, export_path, nodes, evidence, format,
model_pgmpy):
import subprocess, os
if format == "pomegranate":
# immutable index of variables
index = {k: v for v, k in enumerate(nodes)}
inverse_index = {k: v for k, v in enumerate(nodes)}
if format == "pgmpy":
from pgmpy.readwrite import UAIWriter
writer = UAIWriter(model_pgmpy)
domain = writer.get_domain()
domain = sorted(domain.items(), key=lambda x: (x[1], x[0]))
f = open(export_path, "w")
# call DAOOPT and capture its output from file
mydir = os.getcwd()
cmd = [mydir + "/daoopt", "-f", uai_path, "-e", evid_path]
FNULL = open(os.devnull, 'w')
result = subprocess.run(cmd, stdout=subprocess.PIPE, stderr=FNULL)
f.write(result.stdout.decode('utf-8'))
f = open(export_path, "r")
# f = open( os.getcwd() + "/pomegranate.uai.MPE", "r" )
# For example, an input model with 3 binary variables may have a solution line: 3 0 1 0
try:
assignments = [
int(i)
for i in [i for i in f.read().split('\n') if i][-1].split()[3:None]
]
if len(assignments) != len(nodes):
return -1
except ValueError:
return -1
result = {}
for i, v in enumerate(assignments):
if format == "pomegranate":
result[inverse_index[i]] = v
if format == "pgmpy":
result[domain[i][0]] = v
# remove keys in evidence from solution
for k in evidence.keys():
result.pop(k, None)
# return sorted( result.items(), key=lambda x: (x[0], x[1]) )
return result
class TestUAIWriter(unittest.TestCase):
def setUp(self):
self.maxDiff = None
edges = [['family-out', 'dog-out'], ['bowel-problem', 'dog-out'],
['family-out', 'light-on'], ['dog-out', 'hear-bark']]
cpds = {
'bowel-problem': np.array([[0.01], [0.99]]),
'dog-out': np.array([[0.99, 0.01, 0.97, 0.03],
[0.9, 0.1, 0.3, 0.7]]),
'family-out': np.array([[0.15], [0.85]]),
'hear-bark': np.array([[0.7, 0.3], [0.01, 0.99]]),
'light-on': np.array([[0.6, 0.4], [0.05, 0.95]])
}
states = {
'bowel-problem': ['true', 'false'],
'dog-out': ['true', 'false'],
'family-out': ['true', 'false'],
'hear-bark': ['true', 'false'],
'light-on': ['true', 'false']
}
parents = {
'bowel-problem': [],
'dog-out': ['bowel-problem', 'family-out'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']
}
self.bayesmodel = BayesianModel(edges)
tabular_cpds = []
for var, values in cpds.items():
cpd = TabularCPD(var,
len(states[var]),
values,
evidence=parents[var],
evidence_card=[
len(states[evidence_var])
for evidence_var in parents[var]
])
tabular_cpds.append(cpd)
self.bayesmodel.add_cpds(*tabular_cpds)
self.bayeswriter = UAIWriter(self.bayesmodel)
edges = {('var_0', 'var_1'), ('var_0', 'var_2'), ('var_1', 'var_2')}
self.markovmodel = MarkovModel(edges)
tables = [(['var_0', 'var_1'], ['4.000', '2.400', '1.000', '0.000']),
(['var_0', 'var_1', 'var_2'], [
'2.2500', '3.2500', '3.7500', '0.0000', '0.0000',
'10.0000', '1.8750', '4.0000', '3.3330', '2.0000',
'2.0000', '3.4000'
])]
domain = {'var_1': '2', 'var_2': '3', 'var_0': '2'}
factors = []
for table in tables:
variables = table[0]
cardinality = [int(domain[var]) for var in variables]
values = list(map(float, table[1]))
factor = Factor(variables, cardinality, values)
factors.append(factor)
self.markovmodel.add_factors(*factors)
self.markovwriter = UAIWriter(self.markovmodel)
def test_bayes_model(self):
self.expected_bayes_file = """BAYES
5
2 2 2 2 2
5
1 0
3 2 0 1
1 2
2 1 3
2 2 4
2
0.01 0.99
8
0.99 0.01 0.97 0.03 0.9 0.1 0.3 0.7
2
0.15 0.85
4
0.7 0.3 0.01 0.99
4
0.6 0.4 0.05 0.95"""
self.assertEqual(str(self.bayeswriter.__str__()),
str(self.expected_bayes_file))
def test_markov_model(self):
self.expected_markov_file = """MARKOV
3
2 2 3
2
2 0 1
3 0 1 2
4
4.0 2.4 1.0 0.0
12
2.25 3.25 3.75 0.0 0.0 10.0 1.875 4.0 3.333 2.0 2.0 3.4"""
self.assertEqual(str(self.markovwriter.__str__()),
str(self.expected_markov_file))
class TestUAIWriter(unittest.TestCase):
def setUp(self):
self.maxDiff = None
edges = [['family-out', 'dog-out'],
['bowel-problem', 'dog-out'],
['family-out', 'light-on'],
['dog-out', 'hear-bark']]
cpds = {'bowel-problem': np.array([[0.01],
[0.99]]),
'dog-out': np.array([[0.99, 0.01, 0.97, 0.03],
[0.9, 0.1, 0.3, 0.7]]),
'family-out': np.array([[0.15],
[0.85]]),
'hear-bark': np.array([[0.7, 0.3],
[0.01, 0.99]]),
'light-on': np.array([[0.6, 0.4],
[0.05, 0.95]])}
states = {'bowel-problem': ['true', 'false'],
'dog-out': ['true', 'false'],
'family-out': ['true', 'false'],
'hear-bark': ['true', 'false'],
'light-on': ['true', 'false']}
parents = {'bowel-problem': [],
'dog-out': ['bowel-problem', 'family-out'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']}
self.bayesmodel = BayesianModel(edges)
tabular_cpds = []
for var, values in cpds.items():
cpd = TabularCPD(var, len(states[var]), values,
evidence=parents[var],
evidence_card=[len(states[evidence_var])
for evidence_var in parents[var]])
tabular_cpds.append(cpd)
self.bayesmodel.add_cpds(*tabular_cpds)
self.bayeswriter = UAIWriter(self.bayesmodel)
edges = {('var_0', 'var_1'), ('var_0', 'var_2'), ('var_1', 'var_2')}
self.markovmodel = MarkovModel(edges)
tables = [(['var_0', 'var_1'],
['4.000', '2.400', '1.000', '0.000']),
(['var_0', 'var_1', 'var_2'],
['2.2500', '3.2500', '3.7500', '0.0000', '0.0000', '10.0000',
'1.8750', '4.0000', '3.3330', '2.0000', '2.0000', '3.4000'])]
domain = {'var_1': '2', 'var_2': '3', 'var_0': '2'}
factors = []
for table in tables:
variables = table[0]
cardinality = [int(domain[var]) for var in variables]
values = list(map(float, table[1]))
factor = DiscreteFactor(variables, cardinality, values)
factors.append(factor)
self.markovmodel.add_factors(*factors)
self.markovwriter = UAIWriter(self.markovmodel)
def test_bayes_model(self):
self.expected_bayes_file = """BAYES
5
2 2 2 2 2
5
1 0
3 2 0 1
1 2
2 1 3
2 2 4
2
0.01 0.99
8
0.99 0.01 0.97 0.03 0.9 0.1 0.3 0.7
2
0.15 0.85
4
0.7 0.3 0.01 0.99
4
0.6 0.4 0.05 0.95"""
self.assertEqual(str(self.bayeswriter.__str__()), str(self.expected_bayes_file))
def test_markov_model(self):
self.expected_markov_file = """MARKOV
3
2 2 3
2
2 0 1
3 0 1 2
4
4.0 2.4 1.0 0.0
12
2.25 3.25 3.75 0.0 0.0 10.0 1.875 4.0 3.333 2.0 2.0 3.4"""
self.assertEqual(str(self.markovwriter.__str__()), str(self.expected_markov_file))
evidence = model.initial_random_evidence()
# evidence = {'differenziazione': 1, 'loss 17': 2, 'FISH': 3, 'c erbB 2': 1, 'ki67': 3, 'morfologia': 1, 'pM': 1}
# evidence = {}
mpe_pgmpy_ve = model.mpe_query(evidence, algorithm="Variable Elimination")
# MPE external solver
path = "../export/"
# model.export_model_to_uai( path + "my.uai", format="DAOOPT" )
model.export_evidence_to_uai(path + "evidence.uai.evid",
evidence,
format="pgmpy")
from pgmpy.readwrite import UAIWriter, UAIReader
writer = UAIWriter(model.model_pgmpy)
writer.write_uai(path + "pgmpy.uai")
# reader = UAIReader( path + 'pgmpy.uai' )
# model = reader.get_model()
# mpe_daoopt_mine = bn.daoopt_solver( path + "my.uai" , path + "my.uai.evid", path + "sol_my.txt", model.node_names, evidence )
mpe_daoopt_pgmpy = bn.daoopt_solver(path + "pgmpy.uai",
path + "evidence.uai.evid",
path + "sol_pgmpy.txt",
model.node_names, evidence, "pgmpy",
model.model_pgmpy)
print("pgmpy\n", sorted(mpe_pgmpy_ve.items(), key=lambda x: (x[0], x[1])))
# print("DAOOPT my UAI", mpe_daoopt_mine)
print("DAOOPT pgmpy UAI\n",
sorted(mpe_daoopt_pgmpy.items(), key=lambda x: (x[0], x[1])))
def setUp(self):
self.maxDiff = None
variables = [
"kid",
"bowel-problem",
"dog-out",
"family-out",
"hear-bark",
"light-on",
]
edges = [
["family-out", "dog-out"],
["bowel-problem", "dog-out"],
["family-out", "light-on"],
["dog-out", "hear-bark"],
]
cpds = {
"kid": np.array([[0.3], [0.7]]),
"bowel-problem": np.array([[0.01], [0.99]]),
"dog-out": np.array([[0.99, 0.01, 0.97, 0.03], [0.9, 0.1, 0.3, 0.7]]),
"family-out": np.array([[0.15], [0.85]]),
"hear-bark": np.array([[0.7, 0.3], [0.01, 0.99]]),
"light-on": np.array([[0.6, 0.4], [0.05, 0.95]]),
}
states = {
"kid": ["true", "false"],
"bowel-problem": ["true", "false"],
"dog-out": ["true", "false"],
"family-out": ["true", "false"],
"hear-bark": ["true", "false"],
"light-on": ["true", "false"],
}
parents = {
"kid": [],
"bowel-problem": [],
"dog-out": ["bowel-problem", "family-out"],
"family-out": [],
"hear-bark": ["dog-out"],
"light-on": ["family-out"],
}
self.bayesmodel = BayesianModel()
self.bayesmodel.add_nodes_from(variables)
self.bayesmodel.add_edges_from(edges)
tabular_cpds = []
for var, values in cpds.items():
cpd = TabularCPD(
var,
len(states[var]),
values,
evidence=parents[var],
evidence_card=[
len(states[evidence_var]) for evidence_var in parents[var]
],
)
tabular_cpds.append(cpd)
self.bayesmodel.add_cpds(*tabular_cpds)
self.bayeswriter = UAIWriter(self.bayesmodel)
edges = {("var_0", "var_1"), ("var_0", "var_2"), ("var_1", "var_2")}
self.markovmodel = MarkovModel(edges)
tables = [
(["var_0", "var_1"], ["4.000", "2.400", "1.000", "0.000"]),
(
["var_0", "var_1", "var_2"],
[
"2.2500",
"3.2500",
"3.7500",
"0.0000",
"0.0000",
"10.0000",
"1.8750",
"4.0000",
"3.3330",
"2.0000",
"2.0000",
"3.4000",
],
),
]
domain = {"var_1": "2", "var_2": "3", "var_0": "2"}
factors = []
for table in tables:
variables = table[0]
cardinality = [int(domain[var]) for var in variables]
values = list(map(float, table[1]))
factor = DiscreteFactor(variables, cardinality, values)
factors.append(factor)
self.markovmodel.add_factors(*factors)
self.markovwriter = UAIWriter(self.markovmodel)
class TestUAIWriter(unittest.TestCase):
def setUp(self):
self.maxDiff = None
variables = [
"kid",
"bowel-problem",
"dog-out",
"family-out",
"hear-bark",
"light-on",
]
edges = [
["family-out", "dog-out"],
["bowel-problem", "dog-out"],
["family-out", "light-on"],
["dog-out", "hear-bark"],
]
cpds = {
"kid": np.array([[0.3], [0.7]]),
"bowel-problem": np.array([[0.01], [0.99]]),
"dog-out": np.array([[0.99, 0.01, 0.97, 0.03], [0.9, 0.1, 0.3, 0.7]]),
"family-out": np.array([[0.15], [0.85]]),
"hear-bark": np.array([[0.7, 0.3], [0.01, 0.99]]),
"light-on": np.array([[0.6, 0.4], [0.05, 0.95]]),
}
states = {
"kid": ["true", "false"],
"bowel-problem": ["true", "false"],
"dog-out": ["true", "false"],
"family-out": ["true", "false"],
"hear-bark": ["true", "false"],
"light-on": ["true", "false"],
}
parents = {
"kid": [],
"bowel-problem": [],
"dog-out": ["bowel-problem", "family-out"],
"family-out": [],
"hear-bark": ["dog-out"],
"light-on": ["family-out"],
}
self.bayesmodel = BayesianModel()
self.bayesmodel.add_nodes_from(variables)
self.bayesmodel.add_edges_from(edges)
tabular_cpds = []
for var, values in cpds.items():
cpd = TabularCPD(
var,
len(states[var]),
values,
evidence=parents[var],
evidence_card=[
len(states[evidence_var]) for evidence_var in parents[var]
],
)
tabular_cpds.append(cpd)
self.bayesmodel.add_cpds(*tabular_cpds)
self.bayeswriter = UAIWriter(self.bayesmodel)
edges = {("var_0", "var_1"), ("var_0", "var_2"), ("var_1", "var_2")}
self.markovmodel = MarkovModel(edges)
tables = [
(["var_0", "var_1"], ["4.000", "2.400", "1.000", "0.000"]),
(
["var_0", "var_1", "var_2"],
[
"2.2500",
"3.2500",
"3.7500",
"0.0000",
"0.0000",
"10.0000",
"1.8750",
"4.0000",
"3.3330",
"2.0000",
"2.0000",
"3.4000",
],
),
]
domain = {"var_1": "2", "var_2": "3", "var_0": "2"}
factors = []
for table in tables:
variables = table[0]
cardinality = [int(domain[var]) for var in variables]
values = list(map(float, table[1]))
factor = DiscreteFactor(variables, cardinality, values)
factors.append(factor)
self.markovmodel.add_factors(*factors)
self.markovwriter = UAIWriter(self.markovmodel)
def test_bayes_model(self):
self.expected_bayes_file = """BAYES
6
2 2 2 2 2 2
6
1 0
3 2 0 1
1 2
2 1 3
1 4
2 2 5
2
0.01 0.99
8
0.99 0.01 0.97 0.03 0.9 0.1 0.3 0.7
2
0.15 0.85
4
0.7 0.3 0.01 0.99
2
0.3 0.7
4
0.6 0.4 0.05 0.95"""
self.assertEqual(str(self.bayeswriter.__str__()), str(self.expected_bayes_file))
def test_markov_model(self):
self.expected_markov_file = """MARKOV
3
2 2 3
2
2 0 1
3 0 1 2
4
4.0 2.4 1.0 0.0
12
2.25 3.25 3.75 0.0 0.0 10.0 1.875 4.0 3.333 2.0 2.0 3.4"""
self.assertEqual(
str(self.markovwriter.__str__()), str(self.expected_markov_file)
)
