def setUp(self):
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': ['family-out', 'bowel-problem'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']
}
properties = {
'bowel-problem': ['position = (335, 99)'],
'dog-out': ['position = (300, 195)'],
'family-out': ['position = (257, 99)'],
'hear-bark': ['position = (296, 268)'],
'light-on': ['position = (218, 195)']
}
self.model = BayesianModel(edges)
tabular_cpds = []
for var in sorted(cpds.keys()):
values = cpds[var]
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.model.add_cpds(*tabular_cpds)
for node, properties in properties.items():
for prop in properties:
prop_name, prop_value = map(lambda t: t.strip(),
prop.split('='))
self.model.node[node][prop_name] = prop_value
self.writer = BIFWriter(model=self.model)
def create_BN_model(data):
#structure learning
print("Structure learning")
start_time = datetime.now()
print("Start time: ", start_time)
#DECOMMENT TO CREATE A MODEL WITH THE HILL CLIMB ALGORITHM
hc = HillClimbSearch(data)
best_model = hc.estimate()
print(best_model.edges())
edges = best_model.edges()
model = BayesianModel(edges)
print('Fitting the model...')
# Evaluation of cpds using Maximum Likelihood Estimation
model.fit(data)
end_time = datetime.now()
print("End time: ", end_time)
model_write = BIFWriter(model)
model_write.write_bif('model_pgmpy.bif')
if model.check_model():
print(
"Your network structure and CPD's are correctly defined. The probabilities in the columns sum to 1. Hill Climb worked fine!"
)
else:
print("not good")
return (model, end_time - start_time)
def saveModel(model: BayesianModel, file_name: str) -> bool:
"""
A function used to save the given model to BIF file format.
"""
from pgmpy.readwrite import BIFReader, BIFWriter
writer = BIFWriter(model)
try:
writer.write_bif(filename=file_name)
except OSError:
print("Permission denied: Saving file to: ", file_name)
return False
else:
print("Successfully save model to: ", file_name)
return True
def setUp(self):
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': ['family-out', 'bowel-problem'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']}
properties = {'bowel-problem': ['position = (335, 99)'],
'dog-out': ['position = (300, 195)'],
'family-out': ['position = (257, 99)'],
'hear-bark': ['position = (296, 268)'],
'light-on': ['position = (218, 195)']}
self.model = BayesianModel(edges)
tabular_cpds = []
for var in sorted(cpds.keys()):
values = cpds[var]
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.model.add_cpds(*tabular_cpds)
for node, properties in properties.items():
for prop in properties:
prop_name, prop_value = map(lambda t: t.strip(), prop.split('='))
self.model.node[node][prop_name] = prop_value
self.writer = BIFWriter(model=self.model)
class TestBIFWriter(unittest.TestCase):
def setUp(self):
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': ['family-out', 'bowel-problem'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']
}
properties = {
'bowel-problem': ['position = (335, 99)'],
'dog-out': ['position = (300, 195)'],
'family-out': ['position = (257, 99)'],
'hear-bark': ['position = (296, 268)'],
'light-on': ['position = (218, 195)']
}
self.model = BayesianModel(edges)
tabular_cpds = []
for var in sorted(cpds.keys()):
values = cpds[var]
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.model.add_cpds(*tabular_cpds)
for node, properties in properties.items():
for prop in properties:
prop_name, prop_value = map(lambda t: t.strip(),
prop.split('='))
self.model.node[node][prop_name] = prop_value
self.writer = BIFWriter(model=self.model)
def test_str(self):
self.expected_string = """network unknown {
}
variable bowel-problem {
type discrete [ 2 ] { bowel-problem_0, bowel-problem_1 };
property position = (335, 99) ;
}
variable dog-out {
type discrete [ 2 ] { dog-out_0, dog-out_1 };
property position = (300, 195) ;
}
variable family-out {
type discrete [ 2 ] { family-out_0, family-out_1 };
property position = (257, 99) ;
}
variable hear-bark {
type discrete [ 2 ] { hear-bark_0, hear-bark_1 };
property position = (296, 268) ;
}
variable light-on {
type discrete [ 2 ] { light-on_0, light-on_1 };
property position = (218, 195) ;
}
probability ( bowel-problem ) {
table 0.01, 0.99 ;
}
probability ( dog-out | bowel-problem, family-out ) {
table 0.99, 0.01, 0.97, 0.03, 0.9, 0.1, 0.3, 0.7 ;
}
probability ( family-out ) {
table 0.15, 0.85 ;
}
probability ( hear-bark | dog-out ) {
table 0.7, 0.3, 0.01, 0.99 ;
}
probability ( light-on | family-out ) {
table 0.6, 0.4, 0.05, 0.95 ;
}
"""
self.maxDiff = None
self.assertEqual(self.writer.__str__(), self.expected_string)
class TestBIFWriter(unittest.TestCase):
def setUp(self):
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': ['family-out', 'bowel-problem'],
'family-out': [],
'hear-bark': ['dog-out'],
'light-on': ['family-out']}
properties = {'bowel-problem': ['position = (335, 99)'],
'dog-out': ['position = (300, 195)'],
'family-out': ['position = (257, 99)'],
'hear-bark': ['position = (296, 268)'],
'light-on': ['position = (218, 195)']}
self.model = BayesianModel(edges)
tabular_cpds = []
for var in sorted(cpds.keys()):
values = cpds[var]
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.model.add_cpds(*tabular_cpds)
for node, properties in properties.items():
for prop in properties:
prop_name, prop_value = map(lambda t: t.strip(), prop.split('='))
self.model.node[node][prop_name] = prop_value
self.writer = BIFWriter(model=self.model)
def test_str(self):
self.expected_string = """network unknown {
}
variable bowel-problem {
type discrete [ 2 ] { bowel-problem_0, bowel-problem_1 };
property position = (335, 99) ;
}
variable dog-out {
type discrete [ 2 ] { dog-out_0, dog-out_1 };
property position = (300, 195) ;
}
variable family-out {
type discrete [ 2 ] { family-out_0, family-out_1 };
property position = (257, 99) ;
}
variable hear-bark {
type discrete [ 2 ] { hear-bark_0, hear-bark_1 };
property position = (296, 268) ;
}
variable light-on {
type discrete [ 2 ] { light-on_0, light-on_1 };
property position = (218, 195) ;
}
probability ( bowel-problem ) {
table 0.01, 0.99 ;
}
probability ( dog-out | bowel-problem, family-out ) {
table 0.99, 0.01, 0.97, 0.03, 0.9, 0.1, 0.3, 0.7 ;
}
probability ( family-out ) {
table 0.15, 0.85 ;
}
probability ( hear-bark | dog-out ) {
table 0.7, 0.3, 0.01, 0.99 ;
}
probability ( light-on | family-out ) {
table 0.6, 0.4, 0.05, 0.95 ;
}
"""
self.maxDiff = None
self.assertEqual(self.writer.__str__(), self.expected_string)
def save_model(model, fpath):
writer = BIFWriter(model)
writer.write_bif(filename=fpath)
def time_pathfinder_write(self):
BIFWriter(self.pathfinder).write_bif('\tmp')
def time_munin_write(self):
BIFWriter(self.munin).write_bif('\tmp')
def time_asia_write(self):
BIFWriter(self.asia).write_bif('\tmp')