def setUp(self): # Create BayesNet self.bn = BayesianNetwork() # Create Nodes weather0 = DiscreteNode("Weather0", ["Sun", "Rain"]) weather = DiscreteNode("Weather", ["Sun", "Rain"]) ice_cream_eaten = DiscreteNode("Ice Cream Eaten", [True, False]) # Add nodes self.bn.add_node(weather0) self.bn.add_node(weather) self.bn.add_node(ice_cream_eaten) # Add edges self.bn.add_edge(weather, ice_cream_eaten) self.bn.add_edge(weather0, weather) # Set probabilities cpt_weather0 = numpy.array([.6, .4]) weather0.set_probability_table(cpt_weather0, [weather0]) cpt_weather = numpy.array([[.7, .5], [.3, .5]]) weather.set_probability_table(cpt_weather, [weather0, weather]) ice_cream_eaten.set_probability(.9, [(ice_cream_eaten, True), (weather, "Sun")]) ice_cream_eaten.set_probability(.1, [(ice_cream_eaten, False), (weather, "Sun")]) ice_cream_eaten.set_probability(.2, [(ice_cream_eaten, True), (weather, "Rain")]) ice_cream_eaten.set_probability(.8, [(ice_cream_eaten, False), (weather, "Rain")])
def test_complicated_multi(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node" , [True, False,"noIdea"]) cpt1 = numpy.array([2,3]) cpt2 = numpy.array([5,7,9]) n1.set_probability_table(cpt1,[n1]) n2.set_probability_table(cpt2,[n2]) c3 = n1.get_cpd().multiplication(n2.get_cpd()) c3 = n1.get_cpd().multiplication(c3) cptN = numpy.array([[20, 28, 36],[45, 63, 81]]) numpy.testing.assert_array_equal(c3.table,cptN)
def test_complicated_multi(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node", [True, False, "noIdea"]) cpt1 = numpy.array([2, 3]) cpt2 = numpy.array([5, 7, 9]) n1.set_probability_table(cpt1, [n1]) n2.set_probability_table(cpt2, [n2]) c3 = n1.get_cpd().multiplication(n2.get_cpd()) c3 = n1.get_cpd().multiplication(c3) cptN = numpy.array([[20, 28, 36], [45, 63, 81]]) numpy.testing.assert_array_equal(c3.table, cptN)
def test_easy_values(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node" , [True, False]) cpt1 = numpy.array([2,3]) cpt2 = numpy.array([5,7]) n1.set_probability_table(cpt1,[n1]) n2.set_probability_table(cpt2,[n2]) s = n1.get_cpd().multiplication(n2.get_cpd()) cptN = numpy.array([[10,14],[15,21]]) numpy.testing.assert_array_equal(s.table,cptN) self.assertEqual(s.variables[0],n1)
def test_easy_values(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node", [True, False]) cpt1 = numpy.array([2, 3]) cpt2 = numpy.array([5, 7]) n1.set_probability_table(cpt1, [n1]) n2.set_probability_table(cpt2, [n2]) s = n1.get_cpd().multiplication(n2.get_cpd()) cptN = numpy.array([[10, 14], [15, 21]]) numpy.testing.assert_array_equal(s.table, cptN) self.assertEqual(s.variables[0], n1)
def test_easy_marginalize(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node" , [True, False, "other"]) cpt1 = numpy.array([2,3]) cpt2 = numpy.array([5,7,3]) n1.set_probability_table(cpt1,[n1]) n2.set_probability_table(cpt2,[n2]) s = n1.get_cpd().multiplication(n2.get_cpd()) s =s.marginalization(n2) print s.table cptN = numpy.array([30,45]) numpy.testing.assert_array_equal(s.table,cptN) self.assertEqual(s.variables[0],n1)
def test_easy_marginalize(self): n1 = DiscreteNode("Some Node", [True, False]) n2 = DiscreteNode("Second Node", [True, False, "other"]) cpt1 = numpy.array([2, 3]) cpt2 = numpy.array([5, 7, 3]) n1.set_probability_table(cpt1, [n1]) n2.set_probability_table(cpt2, [n2]) s = n1.get_cpd().multiplication(n2.get_cpd()) s = s.marginalization(n2) print s.table cptN = numpy.array([30, 45]) numpy.testing.assert_array_equal(s.table, cptN) self.assertEqual(s.variables[0], n1)
def setUp(self): # Create BayesNet self.bn = BayesianNetwork() # Create Nodes weather0 = DiscreteNode("Weather0", ["Sun", "Rain"]) weather = DiscreteNode("Weather", ["Sun", "Rain"]) ice_cream_eaten = DiscreteNode("Ice Cream Eaten", [True, False]) # Add nodes self.bn.add_node(weather0) self.bn.add_node(weather) self.bn.add_node(ice_cream_eaten) # Add edges self.bn.add_edge(weather, ice_cream_eaten) self.bn.add_edge(weather0, weather) # Set probabilities cpt_weather0 = numpy.array([0.6, 0.4]) weather0.set_probability_table(cpt_weather0, [weather0]) cpt_weather = numpy.array([[0.7, 0.5], [0.3, 0.5]]) weather.set_probability_table(cpt_weather, [weather0, weather]) ice_cream_eaten.set_probability(0.9, [(ice_cream_eaten, True), (weather, "Sun")]) ice_cream_eaten.set_probability(0.1, [(ice_cream_eaten, False), (weather, "Sun")]) ice_cream_eaten.set_probability(0.2, [(ice_cream_eaten, True), (weather, "Rain")]) ice_cream_eaten.set_probability(0.8, [(ice_cream_eaten, False), (weather, "Rain")])
alarm = DiscreteNode("Alarm", ["Ringing", "Silent","Kaputt"]) earthquake = DiscreteNode("Earthquake", ["Shaking", "Calm"]) john_calls = DiscreteNode("John calls", ["Calling", "Not Calling"]) mary_calls = DiscreteNode("Mary calls", ["Calling", "Not Calling"]) bn.add_node(burglary) bn.add_node(alarm) bn.add_node(earthquake) bn.add_node(john_calls) bn.add_node(mary_calls) bn.add_edge(burglary,alarm) bn.add_edge(earthquake, alarm) bn.add_edge(alarm, john_calls) bn.add_edge(alarm, mary_calls) burglary.set_probability(0.2,[(burglary,"Intruder")]) alarm.set_probability(0.1,[(alarm,"Ringing"),(burglary,"Safe"),(earthquake,"Calm")]) cpt = numpy.array([[0.1,0.9],[0.5,0.5],[0.4,0.6]]) john_calls.set_probability_table(cpt, [alarm, john_calls]) print john_calls.is_valid() print alarm.is_valid() bn.draw()
# add Nodes to BayesNet bn.add_node(burglary) bn.add_node(alarm) bn.add_node(earthquake) bn.add_node(john_calls) bn.add_node(baum_calls) # Add edges to show dependencies bn.add_edge(burglary,alarm) bn.add_edge(earthquake, alarm) bn.add_edge(alarm, john_calls) bn.add_edge(alarm, baum_calls) #create probability tables and set them in the node cpt_burglary = numpy.array([0.001,0.999]) burglary.set_probability_table(cpt_burglary,[burglary]) cpt_earthquake = numpy.array([0.002,0.998]) earthquake.set_probability_table(cpt_earthquake,[earthquake]) #another possibility to set probabilities alarm.set_probability(0.95,[(alarm,"Ringing"),(burglary,"Intruder"),(earthquake,"Shaking")]) alarm.set_probability(0.05,[(alarm,"Silent"),(burglary,"Intruder"),(earthquake,"Shaking")]) alarm.set_probability(0.29,[(alarm,"Ringing"),(burglary,"Safe"),(earthquake,"Shaking")]) alarm.set_probability(0.71,[(alarm,"Silent"),(burglary,"Safe"),(earthquake,"Shaking")]) alarm.set_probability(0.94,[(alarm,"Ringing"),(burglary,"Intruder"),(earthquake,"Calm")]) alarm.set_probability(0.06,[(alarm,"Silent"),(burglary,"Intruder"),(earthquake,"Calm")]) alarm.set_probability(0.001,[(alarm,"Ringing"),(burglary,"Safe"),(earthquake,"Calm")]) alarm.set_probability(0.999,[(alarm,"Silent"),(burglary,"Safe"),(earthquake,"Calm")]) baum_calls.set_probability(0.9,[(alarm,"Ringing"),(baum_calls,"Calling")])
#-Probability of evidence #-Maximum a-posteriori hypothesis #Construct some simple BayesianNetwork bn = BayesianNetwork() burglary = DiscreteNode("Burglary", ["Intruder", "Safe"]) alarm = DiscreteNode("Alarm", ["Ringing", "Silent", "Destroyed"]) bn.add_node(burglary) bn.add_node(alarm) bn.add_edge(burglary, alarm) #Parametrize the network burglary_cpt = numpy.array([0.2, 0.8]) burglary.set_probability_table(burglary_cpt, [burglary]) alarm_cpt = numpy.array([[0.8, 0.15, 0.05], [0.05, 0.9, 0.05]]) alarm.set_probability_table(alarm_cpt, [burglary, alarm]) #Get some inference object mcmc_ask = MCMC(bn, 5000, transition_model=GibbsTransitionModel()) #Do some Inferences evidence = {burglary: EvEq("Intruder")} print "-------ProbabilityOfEvidence:-------" poe = mcmc_ask.calculate_PoE(evidence) print "p(evidence=Intruder)=" + str(poe) print "Ground truth=0.2\n"
bn = BayesianNetwork() burglary = DiscreteNode("Burglary", ["Intruder", "Safe"]) alarm = DiscreteNode("Alarm", ["Ringing", "Silent", "Kaputt"]) earthquake = DiscreteNode("Earthquake", ["Shaking", "Calm"]) john_calls = DiscreteNode("John calls", ["Calling", "Not Calling"]) mary_calls = DiscreteNode("Mary calls", ["Calling", "Not Calling"]) bn.add_node(burglary) bn.add_node(alarm) bn.add_node(earthquake) bn.add_node(john_calls) bn.add_node(mary_calls) bn.add_edge(burglary, alarm) bn.add_edge(earthquake, alarm) bn.add_edge(alarm, john_calls) bn.add_edge(alarm, mary_calls) burglary.set_probability(0.2, [(burglary, "Intruder")]) alarm.set_probability(0.1, [(alarm, "Ringing"), (burglary, "Safe"), (earthquake, "Calm")]) cpt = numpy.array([[0.1, 0.9], [0.5, 0.5], [0.4, 0.6]]) john_calls.set_probability_table(cpt, [alarm, john_calls]) print john_calls.is_valid() print alarm.is_valid() bn.draw()
dbn = DynamicBayesianNetwork() B0 = BayesianNetwork() twoTBN = TwoTBN(XMLBIF.read("Robot_Localization.xmlbif")) # Configure TwoTBN x0 = twoTBN.get_node("x0") x = twoTBN.get_node("x") door = twoTBN.get_node("door") twoTBN.set_initial_node(x0.name, x.name) # Configure initial distribution x0_init = DiscreteNode( x0.name, ["p0", "p1", "p2", "p3", "p4", "p5", "p6", "p7", "p8", "p9"]) B0.add_node(x0_init) cpt_x0_init = numpy.array([.1, .1, .1, .1, .1, .1, .1, .1, .1, .1]) x0_init.set_probability_table(cpt_x0_init, [x0_init]) dbn.B0 = B0 dbn.twoTBN = twoTBN N = 1000 T = 10 pos = 0 lastPos = 0 evidence = {} def get_evidence_function(): global pos global evidence
#-Maximum a-posteriori hypothesis #Construct some simple BayesianNetwork bn = BayesianNetwork() burglary = DiscreteNode("Burglary", ["Intruder","Safe"]) alarm = DiscreteNode("Alarm", ["Ringing", "Silent","Destroyed"]) bn.add_node(burglary) bn.add_node(alarm) bn.add_edge(burglary,alarm) #Parametrize the network burglary_cpt=numpy.array([0.2,0.8]) burglary.set_probability_table(burglary_cpt, [burglary]) alarm_cpt=numpy.array([[0.8,0.15,0.05],[0.05,0.9,0.05]]) alarm.set_probability_table(alarm_cpt, [burglary,alarm]) #Get some inference object mcmc_ask=MCMC(bn,5000,transition_model=GibbsTransitionModel()) #Do some Inferences evidence={burglary:EvEq("Intruder")} print "-------ProbabilityOfEvidence:-------" poe=mcmc_ask.calculate_PoE(evidence) print "p(evidence=Intruder)="+str(poe) print "Ground truth=0.2\n"