def get_alarm3_prob(bayes_net):
"""Calculate theprobability of the
temperature being hot (T/F) in the
power plant system, given that the
alarm sounds and neither the gauge
nor alarm is faulty."""
# TODO: finish this function
MA = bayes_net.get_node_by_name('main alarm')
B = bayes_net.get_node_by_name('burglar')
A2 = bayes_net.get_node_by_name('alarm 2')
A3 = bayes_net.get_node_by_name('alarm 3')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[A2] = False
engine.evidence[MA] = False
engine.evidence[B] = True
Q = engine.marginal(A3)[0]
index = Q.generate_index([True],range(Q.nDims))
temp_prob = Q[index]
engine2 = EnumerationEngine(bayes_net)
engine2.evidence[A2] = False # A won against B
engine2.evidence[MA] = False # A won against B
engine2.evidence[B] = True # A won against B
Q = engine.marginal(A3)[0]
posterior = Q.table
print "Or:" , posterior.tolist()
return temp_prob
def get_gauge_prob(bayes_net, gauge_hot):
"""Calculate the marginal probability of the gauge showing hot (T/F) in the power plant system."""
# TODO: finish this function
G_node = bayes_net.get_node_by_name('gauge')
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(G_node)[0]
index = Q.generate_index([gauge_hot],range(Q.nDims))
prob = Q[index]
return prob
def get_alarm_prob(bayes_net, alarm_rings):
"""Calculate the marginal probability of the alarm ringing (T/F) in the power plant system."""
# TODO: finish this function
A_node = bayes_net.get_node_by_name('alarm')
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(A_node)[0]
index = Q.generate_index([alarm_rings],range(Q.nDims))
prob = Q[index]
return prob
def get_not_G_given_not_N(bayes_net):
N_node = bayes_net.get_node_by_name('netflix')
G_node = bayes_net.get_node_by_name('graduate')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[N_node] = False
Q = engine.marginal(G_node)[0]
index = Q.generate_index([False], range(Q.nDims))
graduate_prob = Q[index]
return (graduate_prob)
def get_A_given_E(bayes_net):
E_node = bayes_net.get_node_by_name('exercise')
A_node = bayes_net.get_node_by_name('assignment')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[E_node] = True
Q = engine.marginal(A_node)[0]
index = Q.generate_index([True], range(Q.nDims))
assignment_prob = Q[index]
return (assignment_prob)
def get_gauge_prob(bayes_net, gauge_hot):
"""Calculate the marginal
probability of the gauge
showing hot (T/F) in the
power plant system."""
G_node = bayes_net.get_node_by_name("gauge")
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(G_node)[0]
index = Q.generate_index([gauge_hot], range(Q.nDims))
gauge_prob = Q[index]
return gauge_prob
raise NotImplementedError
def get_E_given_G_N(bayes_net):
E_node = bayes_net.get_node_by_name('exercise')
G_node = bayes_net.get_node_by_name('graduate')
N_node = bayes_net.get_node_by_name('netflix')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[G_node] = True
engine.evidence[N_node] = True
#engine.evidence[N_node] = False
Q = engine.marginal(E_node)[0]
index = Q.generate_index([True], range(Q.nDims))
exercise_prob = Q[index]
return (exercise_prob)
def get_C_given_G_E(bayes_net):
C_node = bayes_net.get_node_by_name('club')
G_node = bayes_net.get_node_by_name('graduate')
E_node = bayes_net.get_node_by_name('exercise')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[G_node] = True
engine.evidence[E_node] = True
#engine.evidence[N_node] = False
Q = engine.marginal(C_node)[0]
index = Q.generate_index([True], range(Q.nDims))
club_prob = Q[index]
return (club_prob)
def get_alarm_prob(bayes_net, alarm_rings):
"""Calculate the marginal
probability of the alarm
ringing (T/F) in the
power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(A_node)[0]
index = Q.generate_index([alarm_rings], range(Q.nDims))
alarm_prob = Q[index]
return alarm_prob
raise NotImplementedError
def get_alarm_prob(bayes_net):
"""Calculate the marginal
probability of the alarm
ringing in the
power plant system."""
# TODO: finish this function
#raise NotImplementedError
A_node = bayes_net.get_node_by_name('alarm')
#T_node = bayes_net.get_node_by_name('temperature')
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(A_node)[0]
index = Q.generate_index([True], range(Q.nDims))
alarm_prob = Q[index]
return alarm_prob
def get_mission_prob_goodbooster(bayes_net):
"""Calculate the marginal
probability of the alarm
ringing in the
power plant system."""
# TODO: finish this function
#raise NotImplementedError
mission_node = bayes_net.get_node_by_name('mission')
#T_node = bayes_net.get_node_by_name('temperature')
engine = JunctionTreeEngine(bayes_net)
Q = engine.marginal(mission_node)[0]
index = Q.generate_index([False], range(Q.nDims))
mission_prob = Q[index]
print(mission_prob)
def get_temperature_prob(bayes_net,temp_hot):
"""Calculate the probability of the temperature being hot (T/F) in the power plant system, given that the
alarm sounds and neither the gauge nor alarm is faulty."""
# TODO: finish this function
T_node = bayes_net.get_node_by_name('temperature')
F_A_node = bayes_net.get_node_by_name('faulty alarm')
F_G_node = bayes_net.get_node_by_name('faulty gauge')
A_node = bayes_net.get_node_by_name('alarm')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[A_node] = True # alarm sounds
engine.evidence[F_A_node] = False # alarm is NOT faulty
engine.evidence[F_G_node] = False # gauge is NOT faulty
Q = engine.marginal(T_node)[0]
index = Q.generate_index([temp_hot],range(Q.nDims))
prob = Q[index]
return prob
def get_mission_prob_badbooster(bayes_net):
"""Calculate the conditional probability
of the temperature being hot in the
power plant system, given that the
alarm sounds and neither the gauge
nor alarm is faulty."""
# TODO: finish this function
#raise NotImplementedError
weather_node = bayes_net.get_node_by_name('weather')
mission_node = bayes_net.get_node_by_name('mission')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[weather_node] = False
Q = engine.marginal(mission_node)[0]
index = Q.generate_index([False], range(Q.nDims))
mission_prob = Q[index]
print(mission_prob)
def get_gauge_prob(bayes_net):
"""Calculate the marginal
probability of the gauge
showing hot in the
power plant system."""
# TODO: finish this function
#raise NotImplementedError
G_node = bayes_net.get_node_by_name('gauge')
#T_node = bayes_net.get_node_by_name('temperature')
#F_G_node = bayes_net.get_node_by_name('faulty gauge')
engine = JunctionTreeEngine(bayes_net)
#engine.evidence[T_node] = True
#engine.evidence[F_G_node] = True
Q = engine.marginal(G_node)[0]
index = Q.generate_index([True], range(Q.nDims))
gauge_prob = Q[index]
return gauge_prob
def get_ghost_prob2(bayes_net):
"""Calculate theprobability of the
temperature being hot (T/F) in the
power plant system, given that the
alarm sounds and neither the gauge
nor alarm is faulty."""
# TODO: finish this function
A3 = bayes_net.get_node_by_name('alarm 3')
A2 = bayes_net.get_node_by_name('alarm 2')
G = bayes_net.get_node_by_name('ghost')
engine = JunctionTreeEngine(bayes_net)
engine.evidence[A2] = False
engine.evidence[A3] = True
Q = engine.marginal(G)[0]
index = Q.generate_index([False],range(Q.nDims))
temp_prob = Q[index]
return temp_prob
def get_alarm_prob(bayes_net, alarm_rings):
"""Calculate the marginal probability of the alarm ringing (T/F) in the power plant system."""
# True or False
alarm_rings = alarm_rings
# set up engine
engine = JunctionTreeEngine(bayes_net)
# get nodes
F_A_node = bayes_net.get_node_by_name('faulty alarm')
G_node = bayes_net.get_node_by_name('gauge')
A_node = bayes_net.get_node_by_name('alarm')
# Compute the marginal probability of alarm given no evidence
Q = engine.marginal(A_node)[0]
#Q is a DiscreteDistribution, and so to index into it, we have to use the class' method to create an index
index = Q.generate_index([False], range(Q.nDims))
alarm_prob_false = Q[index]
alarm_prob_true = float(1 - Q[index])
alarm_prob = alarm_prob_true
print "The marginal probability of alarm=true:", alarm_prob_true
# marginal probablility of alarm ringing
return alarm_prob
def get_temperature_prob(bayes_net, temp_hot):
"""Calculate the conditional probability
of the temperature being hot (T/F) in the
power plant system, given that the
alarm sounds and neither the gauge
nor alarm is faulty."""
# TODO: finish this function
# raise NotImplementedError
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
F_G_node = bayes_net.get_node_by_name("faulty gauge")
T_node = bayes_net.get_node_by_name("temperature")
engine = JunctionTreeEngine(bayes_net)
engine.evidence[A_node] = True
engine.evidence[F_G_node] = False
engine.evidence[F_A_node] = False
Q = engine.marginal(T_node)[0]
index = Q.generate_index([temp_hot], range(Q.nDims))
temp_prob = Q[index]
return temp_prob
def get_gauge_prob(bayes_net, gauge_hot):
"""Calculate the marginal probability of the gauge showing hot (T/F) in the power plant system."""
# True or False
gauge_hot = gauge_hot
# Set up engine
engine = JunctionTreeEngine(bayes_net)
# Get nodes
G_node = bayes_net.get_node_by_name('gauge')
F_G_node = bayes_net.get_node_by_name('faulty gauge')
T_node = bayes_net.get_node_by_name('temperature')
# Compute the marginal probability of gauge given no evidence
Q = engine.marginal(G_node)[0]
#Q is a DiscreteDistribution, and so to index into it, we have to use the class' method to create an index
index = Q.generate_index([False], range(Q.nDims))
gauge_prob_false = Q[index]
gauge_prob_true = float(1 - Q[index])
gauge_prob = gauge_prob_true
print "The marginal probability of gauge=true:", gauge_prob
# marginal probablility of gauge showing hot
return gauge_prob
def get_temperature_prob(bayes_net, temp_hot):
"""Calculate theprobability of the temperature being hot (T/F) in the power plant system, given that the alarm sounds and neither the gauge nor alarm is faulty."""
engine = JunctionTreeEngine(bayes_net)
G_node = bayes_net.get_node_by_name('gauge')
F_G_node = bayes_net.get_node_by_name('faulty gauge')
T_node = bayes_net.get_node_by_name('temperature')
A_node = bayes_net.get_node_by_name('alarm')
F_A_node = bayes_net.get_node_by_name('faulty alarm')
# True of False
temp_hot = temp_hot
engine.evidence[F_A_node] = False
engine.evidence[F_G_node] = False
engine.evidence[A_node] = True
Q = engine.marginal(T_node)[0]
index = Q.generate_index([False], range(Q.nDims))
temp_prob_false = Q[index]
temp_prob_true = float(1 - Q[index])
temp_prob = temp_prob_true
print "The probability of temperature=true (hot) given,"
print "F_A = false, F_G = false, and alarm sounds"
print "temp_prob: ", temp_prob
# the probability that the temperature is actually hot, given that the alarm sounds and the alarm and gauge are both working
return temp_prob
