def set_probability_cargo_flight_goodbooster(bayes_net):
"""Set probability distribution for each node in the power plant system."""
paint_node = bayes_net.get_node_by_name("paint")
weather_node = bayes_net.get_node_by_name("weather")
oring_node = bayes_net.get_node_by_name("oring")
shuttle_node = bayes_net.get_node_by_name("shuttle")
booster_node = bayes_net.get_node_by_name("booster")
mission_node = bayes_net.get_node_by_name("mission")
nodes = [paint_node, weather_node, oring_node, shuttle_node, booster_node, mission_node]
# TODO: set the probability distribution for each node
#raise NotImplementedError
weather_distribution = DiscreteDistribution(weather_node)
index = weather_distribution.generate_index([], [])
weather_distribution[index] = [0.1, 0.9]
weather_node.set_dist(weather_distribution)
paint_distribution = DiscreteDistribution(paint_node)
index = paint_distribution.generate_index([], [])
paint_distribution[index] = [0.01, 0.99]
paint_node.set_dist(paint_distribution)
# Before new arrival
#oring_distribution = DiscreteDistribution(oring_node)
#index = oring_distribution.generate_index([], [])
#oring_distribution[index] = [0.75, 0.25]
#oring_node.set_dist(oring_distribution)
# After new arrival
oring_distribution = DiscreteDistribution(oring_node)
index = oring_distribution.generate_index([], [])
oring_distribution[index] = [0.15, 0.85]
oring_node.set_dist(oring_distribution)
dist = zeros([paint_node.size(), weather_node.size(), shuttle_node.size()], dtype=float32)
dist[0, 0, :] = [0.4, 0.6]
dist[0, 1, :] = [0.4, 0.6]
dist[1, 0, :] = [0.85, 0.15]
dist[1, 1, :] = [0.02, 0.98]
shuttle_distribution = ConditionalDiscreteDistribution(nodes = [paint_node, weather_node, shuttle_node], table=dist)
shuttle_node.set_dist(shuttle_distribution)
dist = zeros([weather_node.size(), oring_node.size(), booster_node.size()], dtype=float32)
dist[0, 0, :] = [0.95, 0.05]
dist[0, 1, :] = [0.15, 0.85]
dist[1, 0, :] = [0.95, 0.05]
dist[1, 1, :] = [0.05, 0.95]
booster_distribution = ConditionalDiscreteDistribution(nodes = [weather_node, oring_node, booster_node], table=dist)
booster_node.set_dist(booster_distribution)
dist = zeros([shuttle_node.size(), booster_node.size(), mission_node.size()], dtype=float32)
dist[0, 0, :] = [0.55, 0.45]
dist[0, 1, :] = [0.2, 0.8]
dist[1, 0, :] = [0.2, 0.8]
dist[1, 1, :] = [0.05, 0.95]
mission_distribution = ConditionalDiscreteDistribution(nodes = [shuttle_node, booster_node, mission_node], table=dist)
mission_node.set_dist(mission_distribution)
return bayes_net
def set_game_probability(bayes_net):
A_node = bayes_net.get_node_by_name('A')
B_node = bayes_net.get_node_by_name('B')
C_node = bayes_net.get_node_by_name('C')
AvB_node = bayes_net.get_node_by_name('AvB')
BvC_node = bayes_net.get_node_by_name('BvC')
CvA_node = bayes_net.get_node_by_name('CvA')
# All AvB, BvC, CvA can use the dist below
dist = zeros([4, 4, 3], dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
# A
A_dist = DiscreteDistribution(A_node)
index = A_dist.generate_index([], [])
A_dist[index] = [0.15, 0.45, 0.30, 0.10]
A_node.set_dist(A_dist)
# B
B_dist = DiscreteDistribution(B_node)
index = B_dist.generate_index([], [])
B_dist[index] = [0.15, 0.45, 0.30, 0.10]
B_node.set_dist(B_dist)
# C
C_dist = DiscreteDistribution(C_node)
index = C_dist.generate_index([], [])
C_dist[index] = [0.15, 0.45, 0.30, 0.10]
C_node.set_dist(C_dist)
# AvB
AvB_node = bayes_net.get_node_by_name('AvB')
AvB_dist = ConditionalDiscreteDistribution(
nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_dist)
# BvC
BvC_node = bayes_net.get_node_by_name('BvC')
BvC_dist = ConditionalDiscreteDistribution(
nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_dist)
# CvA
CvA_node = bayes_net.get_node_by_name('CvA')
CvA_dist = ConditionalDiscreteDistribution(
nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_dist)
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# TODO: set the probability distribution for each node
#Set distribution of node FA
FA_distribution = DiscreteDistribution(F_A_node)
index = FA_distribution.generate_index([], [])
FA_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(FA_distribution)
#Set distribution of node T
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([], [])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
#Set distribution of FG given T
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0, :] = [0.95, 0.05]
dist[1, :] = [0.2, 0.8]
FG_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node],
table=dist)
F_G_node.set_dist(FG_distribution)
#set distribution of A given FA and G
dist = zeros([F_A_node.size(),
G_node.size(), A_node.size()],
dtype=float32)
dist[0, 0, :] = [0.9, 0.1]
dist[0, 1, :] = [0.1, 0.9]
dist[1, 0, :] = [0.55, 0.45]
dist[1, 1, :] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(
nodes=[F_A_node, G_node, A_node], table=dist)
A_node.set_dist(A_distribution)
#Set distribution of G given FG and T
dist = zeros([F_G_node.size(),
T_node.size(), G_node.size()],
dtype=float32)
dist[0, 0, :] = [0.95, 0.05]
dist[0, 1, :] = [0.05, 0.95]
dist[1, 0, :] = [0.2, 0.8]
dist[1, 1, :] = [0.8, 0.2]
G_distribution = ConditionalDiscreteDistribution(
nodes=[F_G_node, T_node, G_node], table=dist)
G_node.set_dist(G_distribution)
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([], [])
T_distribution[index] = [0.80, 0.20]
T_node.set_dist(T_distribution)
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0, :] = [0.95, 0.05]
dist[1, :] = [0.20, 0.80]
F_G_distribution = ConditionalDiscreteDistribution(
nodes=[T_node, F_G_node], table=dist)
F_G_node.set_dist(F_G_distribution)
dist = zeros([T_node.size(), F_G_node.size(),
G_node.size()],
dtype=float32)
dist[0, 0, :] = [0.95, 0.05]
dist[0, 1, :] = [0.20, 0.80]
dist[1, 0, :] = [0.05, 0.95]
dist[1, 1, :] = [0.80, 0.20]
G_distribution = ConditionalDiscreteDistribution(
nodes=[T_node, F_G_node, G_node], table=dist)
G_node.set_dist(G_distribution)
F_A_distribution = DiscreteDistribution(F_A_node)
index = F_A_distribution.generate_index([], [])
F_A_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(F_A_distribution)
dist = zeros([G_node.size(), F_A_node.size(),
A_node.size()],
dtype=float32)
dist[0, 0, :] = [0.90, 0.10]
dist[0, 1, :] = [0.55, 0.45]
dist[1, 0, :] = [0.10, 0.90]
dist[1, 1, :] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(
nodes=[G_node, F_A_node, A_node], table=dist)
A_node.set_dist(A_distribution)
bayes_net = BayesNet(nodes)
return bayes_net
# TODO: set the probability distribution for each node
raise NotImplementedError
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# 1
dist = np.zeros(
[F_G_node.size(), T_node.size(),
G_node.size()], dtype=np.float32)
dist[0, 0, :] = [0.05, 0.95]
dist[0, 1, :] = [0.05, 0.95]
dist[1, 0, :] = [0.8, 0.2]
dist[1, 1, :] = [0.8, 0.2]
G_distribution = ConditionalDiscreteDistribution(
nodes=[F_G_node, T_node, G_node], table=dist)
G_node.set_dist(G_distribution)
# 2
F_A_distribution = DiscreteDistribution(F_A_node)
index = F_A_distribution.generate_index([], [])
F_A_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(F_A_distribution)
# 3
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([], [])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
# 4
dist = np.zeros([T_node.size(), F_G_node.size()], dtype=np.float32)
dist[0, :] = [0.95, 0.05]
dist[1, :] = [0.2, 0.8]
F_G_distribution = ConditionalDiscreteDistribution(
nodes=[T_node, F_G_node], table=dist)
F_G_node.set_dist(F_G_distribution)
# 5
dist = np.zeros(
[F_A_node.size(), G_node.size(),
A_node.size()], dtype=np.float32)
dist[0, 0, :] = [0.1, 0.9]
dist[0, 1, :] = [0.1, 0.9]
dist[1, 0, :] = [0.45, 0.55]
dist[1, 1, :] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(
nodes=[F_A_node, G_node, A_node], table=dist)
A_node.set_dist(A_distribution)
return bayes_net
def set_probability_final_exam(bayes_net):
"""Set probability distribution for each node in the power plant system."""
netflix_node = bayes_net.get_node_by_name("netflix")
exercise_node = bayes_net.get_node_by_name("exercise")
assignment_node = bayes_net.get_node_by_name("assignment")
club_node = bayes_net.get_node_by_name("club")
social_node = bayes_net.get_node_by_name("social")
graduate_node = bayes_net.get_node_by_name("graduate")
nodes = [netflix_node, exercise_node, assignment_node, club_node, social_node, graduate_node]
# TODO: set the probability distribution for each node
#raise NotImplementedError
netflix_distribution = DiscreteDistribution(netflix_node)
index = netflix_distribution.generate_index([], [])
netflix_distribution[index] = [0.27, 0.73]
netflix_node.set_dist(netflix_distribution)
exercise_distribution = DiscreteDistribution(exercise_node)
index = exercise_distribution.generate_index([], [])
exercise_distribution[index] = [0.8, 0.2]
exercise_node.set_dist(exercise_distribution)
# Before new arrival
club_distribution = DiscreteDistribution(club_node)
index = club_distribution.generate_index([], [])
club_distribution[index] = [0.64, 0.36]
club_node.set_dist(club_distribution)
dist = zeros([club_node.size(), social_node.size()], dtype=float32) # Note the order of G_node, A_node
dist[0, :] = [0.31, 0.69] # probabilities for A when G is FALSE
dist[1, :] = [0.22, 0.78] # probabilities for A given G is TRUE
socal_distribution = ConditionalDiscreteDistribution(nodes=[club_node, social_node], table=dist)
social_node.set_dist(socal_distribution)
dist = zeros([netflix_node.size(), exercise_node.size(), assignment_node.size()], dtype=float32)
dist[0, 0, :] = [0.07, 0.93]
dist[0, 1, :] = [0.11, 0.89]
dist[1, 0, :] = [0.82, 0.18]
dist[1, 1, :] = [0.56, 0.44]
assignment_distribution = ConditionalDiscreteDistribution(nodes = [netflix_node, exercise_node, assignment_node], table=dist)
assignment_node.set_dist(assignment_distribution)
dist = zeros([assignment_node.size(), social_node.size(), graduate_node.size()], dtype=float32)
dist[0, 0, :] = [0.7, 0.3]
dist[0, 1, :] = [0.91, 0.09]
dist[1, 0, :] = [0.23, 0.77]
dist[1, 1, :] = [0.48, 0.52]
graduate_distribution = ConditionalDiscreteDistribution(nodes = [assignment_node, social_node, graduate_node], table=dist)
graduate_node.set_dist(graduate_distribution)
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
# temperature distribution
temperature_distribution = DiscreteDistribution(T_node)
index = temperature_distribution.generate_index([],[])
temperature_distribution[index] = [0.80,0.20]
T_node.set_dist(temperature_distribution)
# faulty alarm distribution
faulty_alarm_distribution = DiscreteDistribution(F_A_node)
index = faulty_alarm_distribution.generate_index([],[])
faulty_alarm_distribution[index] = [0.85,0.15]
F_A_node.set_dist(faulty_alarm_distribution)
# faulty gauge distribution
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32) #Note the order of temp, Fg
dist[0,:] = [0.95, 0.05]
dist[1,:] = [0.20, 0.80]
faulty_gauge_distribution = ConditionalDiscreteDistribution(nodes=[T_node,F_G_node], table=dist)
F_G_node.set_dist(faulty_gauge_distribution)
# guage distribution
dist = zeros([T_node.size(), F_G_node.size(), G_node.size()], dtype=float32)
dist[0,0,:] = [0.95, 0.05]
dist[0,1,:] = [0.2, 0.8]
dist[1,0,:] = [0.05, 0.95]
dist[1,1,:] = [0.80, 0.20]
gauge_node_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node, G_node], table=dist)
G_node.set_dist(gauge_node_distribution)
# alarm distribution
dist = zeros([G_node.size(), F_A_node.size(), A_node.size()], dtype=float32)
dist[0,0,:] = [0.90, 0.10]
dist[0,1,:] = [0.55, 0.45]
dist[1,0,:] = [0.10, 0.90]
dist[1,1,:] = [0.45, 0.55]
alarm_node_distribution = ConditionalDiscreteDistribution(nodes=[G_node, F_A_node, A_node], table=dist)
A_node.set_dist(alarm_node_distribution)
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# TODO: set the probability distribution for each node
# Gauge reads the correct temperature with 95% probability when it is not faulty and 20% probability when it is faulty
dist = zeros([T_node.size(), F_G_node.size(), G_node.size()], dtype=float32)
dist[1, 1, :] = [0.8, 0.2]
dist[1, 0, :] = [0.05, 0.95]
dist[0, 1, :] = [0.2, 0.8]
dist[0, 0, :] = [0.95, 0.05]
G_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node, G_node], table=dist)
G_node.set_dist(G_distribution)
# Alarm is faulty 15% of the time
F_A_distribution = DiscreteDistribution(F_A_node)
index = F_A_distribution.generate_index([], [])
F_A_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(F_A_distribution)
# Temperature is hot (call this "true") 20% of the time
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([], [])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
# When temp is hot, the gauge is faulty 80% of the time. Otherwise, the gauge is faulty 5% of the time
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0, :] = [0.95, 0.05]
dist[1, :] = [0.2, 0.8]
F_G_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node], table=dist)
F_G_node.set_dist(F_G_distribution)
# Alarm responds correctly to the gauge 55% of the time when the alarm is faulty,
# and it responds correctly to the gauge 90% of the time when the alarm is not faulty.
dist = zeros([G_node.size(), F_A_node.size(), A_node.size()], dtype=float32)
dist[1, 1, :] = [0.45, 0.55]
dist[1, 0, :] = [0.1, 0.9]
dist[0, 1, :] = [0.55, 0.45]
dist[0, 0, :] = [0.9, 0.1]
A_distribution = ConditionalDiscreteDistribution(nodes=[G_node, F_A_node, A_node], table=dist)
A_node.set_dist(A_distribution)
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each
node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# TODO: set the probability distribution for each node
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([],[])
T_distribution[index] = [0.8,0.2]
T_node.set_dist(T_distribution)
F_A_distribution = DiscreteDistribution(F_A_node)
index = F_A_distribution.generate_index([],[])
F_A_distribution[index] = [0.85,0.15]
F_A_node.set_dist(F_A_distribution)
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0,:] = [0.95, 0.05]
dist[1,:] = [0.2, 0.8]
F_G_distribution = ConditionalDiscreteDistribution(nodes=[T_node,F_G_node], table=dist)
F_G_node.set_dist(F_G_distribution)
dist = zeros([G_node.size(), F_A_node.size(), A_node.size()], dtype=float32)
dist[0,0,:] = [0.9, 0.1]
dist[0,1,:] = [0.55, 0.45]
dist[1,0,:] = [0.1, 0.9]
dist[1,1,:] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(nodes=[G_node, F_A_node, A_node], table=dist)
A_node.set_dist(A_distribution)
dist = zeros([T_node.size(), F_G_node.size(), G_node.size()], dtype=float32)
dist[0,0,:] = [0.95, 0.05]
dist[0,1,:] = [0.2, 0.8]
dist[1,0,:] = [0.05, 0.95]
dist[1,1,:] = [0.8, 0.2]
G_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node, G_node], table=dist)
G_node.set_dist(G_distribution)
return bayes_net
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
FA_node = bayes_net.get_node_by_name("faulty alarm")
G_node = bayes_net.get_node_by_name("gauge")
FG_node = bayes_net.get_node_by_name("faulty gauge")
T_node = bayes_net.get_node_by_name("temperature")
#Temparature
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([],[])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
#Faulty Alarm
FA_distribution = DiscreteDistribution(FA_node)
index = FA_distribution.generate_index([],[])
FA_distribution[index] = [0.85, 0.15]
FA_node.set_dist(FA_distribution)
#Alarm
dist = zeros([G_node.size(), FA_node.size() ,A_node.size()], dtype=float32) #Note the order of G_node, A_node
dist[0,0,:] = [0.90, 0.10]
dist[0,1,:] = [0.55, 0.45]
dist[1,0,:] = [0.10, 0.90]
dist[1,1,:] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(nodes=[G_node, FA_node, A_node], table=dist)
A_node.set_dist(A_distribution)
#Faulty Gauge
dist = zeros([T_node.size(), FG_node.size()], dtype=float32) #Note the order of temp, Fg
dist[0,:] = [0.95, 0.05]
dist[1,:] = [0.20, 0.80]
FG_distribution = ConditionalDiscreteDistribution(nodes=[T_node,FG_node], table=dist)
FG_node.set_dist(FG_distribution)
#Gauge
dist = zeros([T_node.size(), FG_node.size() ,G_node.size()], dtype=float32) #Note the order of G_node, A_node
dist[0,0,:] = [0.95, 0.05]
dist[0,1,:] = [0.20, 0.80]
dist[1,0,:] = [0.05, 0.95]
dist[1,1,:] = [0.80, 0.20]
G_distribution = ConditionalDiscreteDistribution(nodes=[T_node, FG_node, G_node], table=dist)
G_node.set_dist(G_distribution)
return bayes_net
raise NotImplementedError
def make_final_net():
A = BayesNode(0, 2, name='A')
B = BayesNode(1, 2, name='B')
C = BayesNode(2, 2, name='C')
D = BayesNode(3, 2, name='D')
E = BayesNode(4, 2, name='E')
A.add_child(C)
C.add_parent(A)
B.add_child(C)
C.add_parent(B)
B.add_child(D)
D.add_parent(B)
C.add_child(E)
E.add_parent(C)
D.add_child(E)
E.add_parent(D)
nodes = [A, B, C, D, E]
net = BayesNet(nodes)
A = net.get_node_by_name('A')
B = net.get_node_by_name('B')
C = net.get_node_by_name('C')
D = net.get_node_by_name('D')
E = net.get_node_by_name('E')
A_distribution = DiscreteDistribution(A)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.4, 0.6]
A.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.8, 0.2]
B.set_dist(B_distribution)
print B, D
dist = zeros([B.size(), D.size()], dtype=float32)
dist[0, :] = [0.87, 0.13]
dist[1, :] = [0.24, 0.76]
D_distribution = ConditionalDiscreteDistribution(nodes=[B, D], table=dist)
D.set_dist(D_distribution)
dist = zeros([A.size(), B.size(), C.size()], dtype=float32)
dist[0, 0, :] = [0.85, 0.15]
dist[0, 1, :] = [0.68, 0.32]
dist[1, 0, :] = [0.16, 0.84]
dist[1, 1, :] = [0.05, 0.95]
C_distribution = ConditionalDiscreteDistribution(nodes=[A, B, C],
table=dist)
C.set_dist(C_distribution)
dist = zeros([C.size(), D.size(), E.size()], dtype=float32)
dist[0, 0, :] = [0.8, 0.2]
dist[0, 1, :] = [0.37, 0.63]
dist[1, 0, :] = [0.08, 0.92]
dist[1, 1, :] = [0.07, 0.93]
E_distribution = ConditionalDiscreteDistribution(nodes=[C, D, E],
table=dist)
E.set_dist(E_distribution)
engine = EnumerationEngine(net)
# engine1.evidence[wear] = False
# engine1.evidence[weap] = True
engine.evidence[A] = True
engine.evidence[C] = True
Q = engine.marginal(E)[0]
index = Q.generate_index([True], range(Q.nDims))
prob = Q[index]
print prob
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
# TODO: fill this out
A = BayesNode(0, 4, name='A')
B = BayesNode(1, 4, name='B')
C = BayesNode(2, 4, name='C')
AB = BayesNode(3, 3, name='AvB')
BC = BayesNode(4, 3, name='BvC')
CA = BayesNode(5, 3, name='CvA')
A.add_child(AB)
A.add_child(CA)
B.add_child(AB)
B.add_child(BC)
C.add_child(BC)
C.add_child(CA)
AB.add_parent(A)
AB.add_parent(B)
BC.add_parent(B)
BC.add_parent(C)
CA.add_parent(A)
CA.add_parent(C)
# Set A distribution
A_distribution = DiscreteDistribution(A)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.15, 0.45, 0.30, 0.10]
A.set_dist(A_distribution)
# Set B distribution
B_distribution = DiscreteDistribution(B)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.15, 0.45, 0.30, 0.10]
B.set_dist(B_distribution)
# Set C distribution
C_distribution = DiscreteDistribution(C)
index = C_distribution.generate_index([], [])
C_distribution[index] = [0.15, 0.45, 0.30, 0.10]
C.set_dist(C_distribution)
# Set distribution of AvB given A and B
dist = zeros([A.size(), B.size(), AB.size()], dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
AB_distribution = ConditionalDiscreteDistribution(nodes=[A, B, AB],
table=dist)
AB.set_dist(AB_distribution)
# Set distribution of BvC given B and C
dist = zeros([B.size(), C.size(), BC.size()], dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
BC_distribution = ConditionalDiscreteDistribution(nodes=[B, C, BC],
table=dist)
BC.set_dist(BC_distribution)
# Set distribution of CA giveen C and A
dist = zeros([C.size(), A.size(), CA.size()], dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
CA_distribution = ConditionalDiscreteDistribution(nodes=[C, A, CA],
table=dist)
CA.set_dist(CA_distribution)
nodes = [A, B, C, AB, BC, CA]
return BayesNet(nodes)
def get_game_network():
# intial setup for problem described above
nodes = []
numberOfNodes = 6
A = 0
B = 1
C = 2
AvB = 3
BvC = 4
CvA = 5
# create the nodes
A_node = BayesNode(0, 4, name='A')
B_node = BayesNode(1, 4, name='B')
C_node = BayesNode(2, 4, name='C')
AvB_node = BayesNode(3, 3, name='AvB')
BvC_node = BayesNode(4, 3, name='BvC')
CvA_node = BayesNode(5, 3, name='CvA')
# setup A Node prior distribution
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.15, 0.45, 0.3, 0.1]
A_node.set_dist(A_distribution)
# setup B Node prior distribution
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.15, 0.45, 0.3, 0.1]
B_node.set_dist(B_distribution)
# setup C Node prior distribution
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([], [])
C_distribution[index] = [0.15, 0.45, 0.3, 0.1]
C_node.set_dist(C_distribution)
# Probabilty Table for Matchup of Two Teams based on Skill Difference
# | skill difference (T2 - T1) | T1 wins | T2 wins | Tie |
# |0 |0.10 |0.10 |0.80|
# |1 |0.20 |0.60 |0.20|
# |2 |0.15 |0.75 |0.10|
# |3 |0.05 |0.90 |0.05|
# |-1 |0.60 |0.20 |0.20|
# |-2 |0.75 |0.15 |0.10|
# |-3 |0.90 |0.05 |0.05|
# setup AvB Node distribution
AvB_distribution = DiscreteDistribution(AvB_node)
dist = zeros([A_node.size(), B_node.size(),
AvB_node.size()],
dtype=float32)
for a in range(A_node.size()):
for b in range(B_node.size()):
if (b - a) == -3:
dist[a, b, :] = [0.90, 0.05, 0.05]
elif (b - a) == -2:
dist[a, b, :] = [0.75, 0.15, 0.10]
elif (b - a) == -1:
dist[a, b, :] = [0.60, 0.20, 0.20]
elif (b - a) == 0:
dist[a, b, :] = [0.10, 0.10, 0.80]
elif (b - a) == 1:
dist[a, b, :] = [0.20, 0.60, 0.20]
elif (b - a) == 2:
dist[a, b, :] = [0.15, 0.75, 0.10]
elif (b - a) == 3:
dist[a, b, :] = [0.05, 0.90, 0.05]
else:
print "ERROR in AvB node setup"
AvB_distribution = ConditionalDiscreteDistribution(
nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
# setup BvC Node distribution
BvC_distribution = DiscreteDistribution(BvC_node)
dist = zeros([B_node.size(), C_node.size(),
BvC_node.size()],
dtype=float32)
for b in range(B_node.size()):
for c in range(C_node.size()):
if (c - b) == -3:
dist[b, c, :] = [0.90, 0.05, 0.05]
elif (c - b) == -2:
dist[b, c, :] = [0.75, 0.15, 0.10]
elif (c - b) == -1:
dist[b, c, :] = [0.60, 0.20, 0.20]
elif (c - b) == 0:
dist[b, c, :] = [0.10, 0.10, 0.80]
elif (c - b) == 1:
dist[b, c, :] = [0.20, 0.60, 0.20]
elif (c - b) == 2:
dist[b, c, :] = [0.15, 0.75, 0.10]
elif (c - b) == 3:
dist[b, c, :] = [0.05, 0.90, 0.05]
else:
print "ERROR in BvC node setup"
BvC_distribution = ConditionalDiscreteDistribution(
nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
# setup CvA Node distribution
CvA_distribution = DiscreteDistribution(CvA_node)
dist = zeros([C_node.size(), A_node.size(),
CvA_node.size()],
dtype=float32)
for c in range(C_node.size()):
for a in range(A_node.size()):
if (a - c) == -3:
dist[c, a, :] = [0.90, 0.05, 0.05]
elif (a - c) == -2:
dist[c, a, :] = [0.75, 0.15, 0.10]
elif (a - c) == -1:
dist[c, a, :] = [0.60, 0.20, 0.20]
elif (a - c) == 0:
dist[c, a, :] = [0.10, 0.10, 0.80]
elif (a - c) == 1:
dist[c, a, :] = [0.20, 0.60, 0.20]
elif (a - c) == 2:
dist[c, a, :] = [0.15, 0.75, 0.10]
elif (a - c) == 3:
dist[c, a, :] = [0.05, 0.90, 0.05]
else:
print "ERROR in CvA node setup"
CvA_distribution = ConditionalDiscreteDistribution(
nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_distribution)
# Setup Network (Parents & Children)
# A
A_node.add_child(AvB_node)
A_node.add_child(CvA_node)
# B
B_node.add_child(AvB_node)
B_node.add_child(BvC_node)
# C
C_node.add_child(BvC_node)
C_node.add_child(CvA_node)
# AvB
AvB_node.add_parent(A_node)
AvB_node.add_parent(B_node)
# BvC
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
# CvA
CvA_node.add_parent(C_node)
CvA_node.add_parent(A_node)
# add the nodes for setting up network
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
return BayesNet(nodes)
def make_exam_net():
H = BayesNode(0, 2, name='H')
G = BayesNode(1, 2, name='G')
B = BayesNode(2, 2, name='B')
O = BayesNode(3, 2, name='O')
D = BayesNode(4, 2, name='D')
C = BayesNode(5, 2, name='C')
H.add_child(B)
B.add_parent(H)
B.add_parent(G)
G.add_child(B)
G.add_child(O)
O.add_parent(G)
O.add_child(D)
O.add_child(C)
D.add_parent(O)
C.add_parent(O)
B.add_child(D)
D.add_parent(B)
nodes = [B, C, D, G, H, O]
net = BayesNet(nodes)
sci = BayesNode(0, 2, name='sci')
inf = BayesNode(1, 2, name='inf')
weap = BayesNode(2, 2, name='weap')
car = BayesNode(3, 2, name='car')
wear = BayesNode(4, 2, name='wear')
sci.add_child(weap)
weap.add_parent(sci)
inf.add_child(weap)
weap.add_parent(inf)
weap.add_child(car)
car.add_parent(weap)
weap.add_child(wear)
wear.add_parent(weap)
nodes1 = [sci, weap, wear, inf, car]
net1 = BayesNet(nodes1)
B = net.get_node_by_name('B')
C = net.get_node_by_name('C')
D = net.get_node_by_name('D')
G = net.get_node_by_name('G')
H = net.get_node_by_name('H')
O = net.get_node_by_name('O')
H_distribution = DiscreteDistribution(H)
index = H_distribution.generate_index([], [])
H_distribution[index] = [0.6, 0.4]
H.set_dist(H_distribution)
G_distribution = DiscreteDistribution(G)
index = G_distribution.generate_index([], [])
G_distribution[index] = [0.75, 0.25]
G.set_dist(G_distribution)
dist = zeros([O.size(), C.size()], dtype=float32)
dist[0, :] = [0.55, 0.45]
dist[1, :] = [0.75, 0.25]
C_distribution = ConditionalDiscreteDistribution(nodes=[O, C], table=dist)
C.set_dist(C_distribution)
dist = zeros([G.size(), O.size()], dtype=float32)
dist[0, :] = [0.55, 0.45]
dist[1, :] = [0.45, 0.55]
O_distribution = ConditionalDiscreteDistribution(nodes=[G, O], table=dist)
O.set_dist(O_distribution)
dist = zeros([B.size(), O.size(), D.size()], dtype=float32)
dist[0, 0, :] = [0.72, 0.28]
dist[0, 1, :] = [0.38, 0.62]
dist[1, 0, :] = [0.85, 0.15]
dist[1, 1, :] = [0.65, 0.35]
D_distribution = ConditionalDiscreteDistribution(nodes=[B, O, D],
table=dist)
D.set_dist(D_distribution)
dist = zeros([H.size(), G.size(), B.size()], dtype=float32)
dist[0, 0, :] = [0.92, 0.08]
dist[0, 1, :] = [0.75, 0.25]
dist[1, 0, :] = [0.55, 0.45]
dist[1, 1, :] = [0.35, 0.65]
B_distribution = ConditionalDiscreteDistribution(nodes=[H, G, B],
table=dist)
B.set_dist(B_distribution)
sci = net1.get_node_by_name('sci')
weap = net1.get_node_by_name('weap')
wear = net1.get_node_by_name('wear')
inf = net1.get_node_by_name('inf')
car = net1.get_node_by_name('car')
sci_distribution = DiscreteDistribution(sci)
index = sci_distribution.generate_index([], [])
sci_distribution[index] = [0.2, 0.8]
sci.set_dist(sci_distribution)
inf_distribution = DiscreteDistribution(inf)
index = inf_distribution.generate_index([], [])
inf_distribution[index] = [0.4, 0.6]
inf.set_dist(inf_distribution)
dist = zeros([sci.size(), inf.size(), weap.size()], dtype=float32)
dist[0, 0, :] = [0.6, 0.4]
dist[0, 1, :] = [0.8, 0.2]
dist[1, 0, :] = [0.1, 0.9]
dist[1, 1, :] = [0.3, 0.7]
weap_distribution = ConditionalDiscreteDistribution(nodes=[sci, inf, weap],
table=dist)
weap.set_dist(weap_distribution)
dist = zeros([weap.size(), wear.size()], dtype=float32)
dist[0, :] = [0.88, 0.12]
dist[1, :] = [0.15, 0.85]
wear_distribution = ConditionalDiscreteDistribution(nodes=[weap, wear],
table=dist)
wear.set_dist(wear_distribution)
dist = zeros([weap.size(), car.size()], dtype=float32)
dist[0, :] = [0.75, 0.25]
dist[1, :] = [0.55, 0.45]
car_distribution = ConditionalDiscreteDistribution(nodes=[weap, car],
table=dist)
car.set_dist(car_distribution)
## engine = JunctionTreeEngine(net)
# engine = EnumerationEngine(net)
## engine.evidence[B] = True
# Q = engine.marginal(C)[0]
# index = Q.generate_index([True], range(Q.nDims))
# prob = Q[index]
# print "Thr ptob of O = T given B = T is ", prob
engine1 = EnumerationEngine(net1)
engine1.evidence[wear] = False
engine1.evidence[weap] = True
# engine1.evidence[sci] = False
Q = engine1.marginal(car)[0]
index = Q.generate_index([True], range(Q.nDims))
prob = Q[index]
print prob
def get_game_network():
"""Create a Bayes Net representation
of the game problem."""
nodes = []
# TODO: fill this out
A_node = BayesNode(0,4,name='A')
B_node = BayesNode(1,4,name='B')
C_node = BayesNode(2,4,name='C')
AvB_node = BayesNode(3,3,name='AvB')
BvC_node = BayesNode(4,3,name='BvC')
CvA_node = BayesNode(5,3,name='CvA')
A_node.add_child(AvB_node)
AvB_node.add_parent(A_node)
B_node.add_child(AvB_node)
AvB_node.add_parent(B_node)
B_node.add_child(BvC_node)
BvC_node.add_parent(B_node)
C_node.add_child(BvC_node)
BvC_node.add_parent(C_node)
C_node.add_child(CvA_node)
CvA_node.add_parent(C_node)
A_node.add_child(CvA_node)
CvA_node.add_parent(A_node)
nodes.append(A_node)
nodes.append(B_node)
nodes.append(C_node)
nodes.append(AvB_node)
nodes.append(BvC_node)
nodes.append(CvA_node)
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([],[])
A_distribution[index] = [0.15,0.45,0.3,0.1]
A_node.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([],[])
B_distribution[index] = [0.15,0.45,0.3,0.1]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([],[])
C_distribution[index] = [0.15,0.45,0.3,0.1]
C_node.set_dist(C_distribution)
dist = zeros([A_node.size(), B_node.size(), AvB_node.size()], dtype=float32)
dist[0,0,:] = [0.1, 0.1, 0.8]
dist[0,1,:] = [0.2, 0.6, 0.2]
dist[0,2,:] = [0.15, 0.75, 0.1]
dist[0,3,:] = [0.05, 0.9, 0.05]
dist[1,0,:] = [0.6, 0.2, 0.2]
dist[1,1,:] = [0.1, 0.1, 0.8]
dist[1,2,:] = [0.2, 0.6, 0.2]
dist[1,3,:] = [0.15, 0.75, 0.1]
dist[2,0,:] = [0.75, 0.15, 0.1]
dist[2,1,:] = [0.6, 0.2, 0.2]
dist[2,2,:] = [0.1, 0.1, 0.8]
dist[2,3,:] = [0.2, 0.6, 0.2]
dist[3,0,:] = [0.9, 0.05, 0.05]
dist[3,1,:] = [0.75, 0.15, 0.1]
dist[3,2,:] = [0.6, 0.2, 0.2]
dist[3,3,:] = [0.1, 0.1, 0.8]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
CvA_distribution = ConditionalDiscreteDistribution(nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_distribution)
print "Printing table"
print type(AvB_node.dist.table)
for i in range(3):
print AvB_node.dist.table[0][0][i]
return BayesNet(nodes)
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
teama_node = BayesNode(0,4,name='A')
teamb_node = BayesNode(1,4,name='B')
teamc_node = BayesNode(2,4,name='C')
matchAvB_node = BayesNode(3,3,name='AvB')
matchBvC_node = BayesNode(4,3,name='BvC')
matchCvA_node = BayesNode(5,3,name='CvA')
teama_node.add_child(matchAvB_node)
matchAvB_node.add_parent(teama_node)
teama_node.add_child(matchCvA_node)
matchCvA_node.add_parent(teama_node)
teamb_node.add_child(matchAvB_node)
matchAvB_node.add_parent(teamb_node)
teamb_node.add_child(matchBvC_node)
matchBvC_node.add_parent(teamb_node)
teamc_node.add_child(matchBvC_node)
matchBvC_node.add_parent(teamc_node)
teamc_node.add_child(matchCvA_node)
matchCvA_node.add_parent(teamc_node)
skill_dist = [0.15,0.45,0.30,0.10]
teama_distribution = DiscreteDistribution(teama_node)
index = teama_distribution.generate_index([],[])
teama_distribution[index] = skill_dist
teama_node.set_dist(teama_distribution)
teamb_distribution = DiscreteDistribution(teamb_node)
index = teamb_distribution.generate_index([],[])
teamb_distribution[index] = skill_dist
teamb_node.set_dist(teamb_distribution)
teamc_distribution = DiscreteDistribution(teamc_node)
index = teamc_distribution.generate_index([],[])
teamc_distribution[index] = skill_dist
teamc_node.set_dist(teamc_distribution)
match_dist = zeros([teama_node.size(), teamb_node.size(), matchAvB_node.size()], dtype=float32)
match_dist[0,0,:] = [0.10, 0.10, 0.80]
match_dist[0,1,:] = [0.20, 0.60, 0.20]
match_dist[0,2,:] = [0.15, 0.75, 0.10]
match_dist[0,3,:] = [0.05, 0.90, 0.05]
match_dist[1,0,:] = [0.60, 0.20, 0.20]
match_dist[1,1,:] = [0.10, 0.10, 0.80]
match_dist[1,2,:] = [0.20, 0.60, 0.20]
match_dist[1,3,:] = [0.15, 0.75, 0.10]
match_dist[2,0,:] = [0.75, 0.15, 0.10]
match_dist[2,1,:] = [0.60, 0.20, 0.20]
match_dist[2,2,:] = [0.10, 0.10, 0.80]
match_dist[2,3,:] = [0.20, 0.60, 0.20]
match_dist[3,0,:] = [0.90, 0.05, 0.05]
match_dist[3,1,:] = [0.75, 0.15, 0.10]
match_dist[3,2,:] = [0.60, 0.20, 0.20]
match_dist[3,3,:] = [0.10, 0.10, 0.80]
matchAvB_distribution = ConditionalDiscreteDistribution(nodes=[teama_node, teamb_node, matchAvB_node], table=match_dist)
matchBvC_distribution = ConditionalDiscreteDistribution(nodes=[teamb_node, teamc_node, matchBvC_node], table=match_dist)
matchCvA_distribution = ConditionalDiscreteDistribution(nodes=[teamc_node, teama_node, matchCvA_node], table=match_dist)
matchAvB_node.set_dist(matchAvB_distribution)
matchBvC_node.set_dist(matchBvC_distribution)
matchCvA_node.set_dist(matchCvA_distribution)
nodes = [teama_node,teamb_node,teamc_node,matchAvB_node,matchBvC_node,matchCvA_node]
return BayesNet(nodes)
def set_probability(bayes_net):
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# Use the following Boolean variables in your implementation:
# A = alarm sounds
# F_A = alarm is faulty
# G = gauge reading (high = True, normal = False)
# F_G = gauge is faulty
# T = actual temperature (high = True, normal = False)
# temperature node distribution
# 0 Index = False Prob
# 1 Index = True Prob
# True = Hot Temp = 20%
# False = Normal Temp = 80%
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([], [])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
# faulty alarm node distribution
# 0 Index = False Prob
# 1 Index = True Prob
# True = alarm is faulty = 15%
# False = alarm is not faulty = 85%
F_A_distribution = DiscreteDistribution(F_A_node)
index = F_A_distribution.generate_index([], [])
F_A_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(F_A_distribution)
# faulty gauge node distribution
# 0 column -- when temp is normal (T = False)
# 1 Index -- when temp is hot (T = True)
# True = faulty alarm, False = not faulty alarm
# when Temp is normal (T=F),
# F_G = false = .95 normal alarm with normal temp
# F_G = true = .05 faulty alarm with normal temp
# when Temp is hot (T=T),
# F_G = false = .20 normal alarm with hot temp
# F_G = true = .80 faulty alarm with hot temp
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0, :] = [0.95, 0.05]
dist[1, :] = [0.20, 0.80]
F_G_distribution = ConditionalDiscreteDistribution(
nodes=[T_node, F_G_node], table=dist)
F_G_node.set_dist(F_G_distribution)
# gauge node distribution
#Temp: hot= T normal = F
#F_G: faulty= T not faulty/normal = F
#G: hot = T normal = F
# Temp F_G P(G=true|Temp,F_G)
# T T 0.20
# T F 0.95
# F T 0.80
# F F 0.05
# Temp = Hot = True
# F_G = Faulty = True
# True, True, True = .20
dist = zeros([T_node.size(), F_G_node.size(),
G_node.size()],
dtype=float32)
dist[1, 1, :] = [0.80, 0.20]
dist[1, 0, :] = [0.05, 0.95]
dist[0, 1, :] = [0.20, 0.80]
dist[0, 0, :] = [0.95, 0.05]
G_distribution = ConditionalDiscreteDistribution(
nodes=[T_node, F_G_node, G_node], table=dist)
G_node.set_dist(G_distribution)
# alarm node distribution
#Gauge: hot= T normal = F
#F_A: faulty= T not faulty/normal = F
#A: sounds = T alarm doesnt sound = F
# Gauge F_A P(A=true|G,F_A)
# T T 0.55
# T F 0.90
# F T 0.45
# F F 0.10
# Gauge = Hot = True
# F_A = Faulty = True
# True, True, True = .55
dist = zeros([G_node.size(), F_A_node.size(),
A_node.size()],
dtype=float32)
dist[1, 1, :] = [0.45, 0.55]
dist[1, 0, :] = [0.10, 0.90]
dist[0, 1, :] = [0.55, 0.45]
dist[0, 0, :] = [0.90, 0.10]
A_distribution = ConditionalDiscreteDistribution(
nodes=[G_node, F_A_node, A_node], table=dist)
A_node.set_dist(A_distribution)
return bayes_net
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodes = []
# TODO: fill this out
# Create Nodes
A = BayesNode(0, 4, name="A")
B_node = BayesNode(1, 4, name="B")
C_node = BayesNode(2, 4, name="C")
AvB = BayesNode(3, 3, name="AvB")
BvC_node = BayesNode(4, 3, name="BvC")
CvA = BayesNode(5, 3, name="CvA")
# A skill
A.add_child(AvB)
A.add_child(CvA)
# B skill
B_node.add_child(AvB)
B_node.add_child(BvC_node)
# C skill
C_node.add_child(BvC_node)
C_node.add_child(CvA)
# AvB outcome
AvB.add_parent(A)
AvB.add_parent(B_node)
# BvC outcome
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
# CvA outcome
CvA.add_parent(C_node)
CvA.add_parent(A)
# Set Probability Distributions
# A, B, C team skill levels
A_distribution = DiscreteDistribution(A)
index = A_distribution.generate_index([],[])
A_distribution[index]=[0.15, 0.45, 0.3, 0.1]
A.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([],[])
B_distribution[index]=[0.15, 0.45, 0.3, 0.1]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([],[])
C_distribution[index]=[0.15, 0.45, 0.3, 0.1]
C_node.set_dist(C_distribution)
# AvB, BvC, CvA game outcomes
dist = zeros([A.size(), B_node.size(), AvB.size()], dtype=float32)
dist[0,0,:] = [0.1, 0.1, 0.8]
dist[0,1,:] = [0.2, 0.6, 0.2]
dist[0,2,:] = [0.15, 0.75, 0.1]
dist[0,3,:] = [0.05, 0.9, 0.05]
dist[1,0,:] = [0.6, 0.2, 0.2]
dist[1,1,:] = [0.1, 0.1, 0.8]
dist[1,2,:] = [0.2, 0.6, 0.2]
dist[1,3,:] = [0.15, 0.75, 0.1]
dist[2,0,:] = [0.75, 0.15, 0.1]
dist[2,1,:] = [0.6, 0.2, 0.2]
dist[2,2,:] = [0.1, 0.1, 0.8]
dist[2,3,:] = [0.2, 0.6, 0.2]
dist[3,0,:] = [0.9, 0.05, 0.05]
dist[3,1,:] = [0.75, 0.15, 0.1]
dist[3,2,:] = [0.6, 0.2, 0.2]
dist[3,3,:] = [0.1, 0.1, 0.8]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A, B_node, AvB], table=dist)
AvB.set_dist(AvB_distribution)
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
CvA_distribution = ConditionalDiscreteDistribution(nodes=[C_node, A, CvA], table=dist)
CvA.set_dist(CvA_distribution)
nodes = [A, B_node, C_node, AvB, BvC_node, CvA]
return BayesNet(nodes)
def set_skill_distribution(node):
distribution = DiscreteDistribution(node)
index = distribution.generate_index([], [])
distribution[index] = [0.15, 0.45, 0.3, 0.1]
node.set_dist(distribution)
return node
def set_probability(bayes_net):
"""Set probability distribution for each node in the power plant system."""
A_node = bayes_net.get_node_by_name("alarm")
F_A_node = bayes_net.get_node_by_name("faulty alarm")
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")
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
# TODO: set the probability distribution for each node
# 1. Gauge Probability Distribution
# T | Fg | P(G | T, Fg)
# F | F | 0.05
# F | T | 0.8
# T | F | 0.95
# T | T | 0.2
dist = zeros([T_node.size(), F_G_node.size(), G_node.size()], dtype=float32)
dist[0,0,:] = [0.95, 0.05]
dist[0,1,:] = [0.2, 0.8]
dist[1,0,:] = [0.05, 0.95]
dist[1,1,:] = [0.8, 0.2]
G_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node, G_node],
table=dist)
G_node.set_dist(G_distribution)
# 2. Faulty Alarm Probability Distribution
# P(Fa) = 0.15
Fa_distribution = DiscreteDistribution(F_A_node)
index = Fa_distribution.generate_index([], [])
Fa_distribution[index] = [0.85, 0.15]
F_A_node.set_dist(Fa_distribution)
# 3. Actual Temperature Probability Distribution
# P(T) = 0.2
T_distribution = DiscreteDistribution(T_node)
index = T_distribution.generate_index([],[])
T_distribution[index] = [0.8, 0.2]
T_node.set_dist(T_distribution)
# 4. Faulty Gauge Probability Distribution
# T | P(Fg | T)
# F | 0.05
# T | 0.8
dist = zeros([T_node.size(), F_G_node.size()], dtype=float32)
dist[0, :] = [.95,0.05]
dist[1, :] = [0.2,0.8]
Fg_distribution = ConditionalDiscreteDistribution(nodes=[T_node, F_G_node], table=dist)
F_G_node.set_dist(Fg_distribution)
# 5. Alarm Probability Distribution
# Fa | G | P(A | Fa, G)
# F | F | 0.1
# F | T | 0.9
# T | F | 0.45
# T | T | 0.55
dist = zeros([F_A_node.size(), G_node.size(), A_node.size()], dtype=float32)
dist[0,0,:] = [0.9, 0.1]
dist[0,1,:] = [0.1, 0.9]
dist[1,0,:] = [0.55, 0.45]
dist[1,1,:] = [0.45, 0.55]
A_distribution = ConditionalDiscreteDistribution(nodes=[F_A_node, G_node, A_node],
table=dist)
A_node.set_dist(A_distribution)
return bayes_net
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodes = []
# TODO: fill this out
A_node = BayesNode(0, 4, name='A')
B_node = BayesNode(1, 4, name='B')
C_node = BayesNode(2, 4, name='C')
AvB_node = BayesNode(3, 3, name='AvB')
BvC_node = BayesNode(4, 3, name='BvC')
CvA_node = BayesNode(5, 3, name='CvA')
# Match A v B
AvB_node.add_parent(A_node)
AvB_node.add_parent(B_node)
A_node.add_child(AvB_node)
B_node.add_child(AvB_node)
# Match B v C
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
B_node.add_child(BvC_node)
C_node.add_child(BvC_node)
# Match C v A
CvA_node.add_parent(C_node)
CvA_node.add_parent(A_node)
C_node.add_child(CvA_node)
A_node.add_child(CvA_node)
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
prior_skill_dist = [0.15, 0.45, 0.3, 0.1]
A_skill_dist = DiscreteDistribution(A_node)
index = A_skill_dist.generate_index([], [])
A_skill_dist[index] = prior_skill_dist
A_node.set_dist(A_skill_dist)
B_skill_dist = DiscreteDistribution(B_node)
index = B_skill_dist.generate_index([], [])
B_skill_dist[index] = prior_skill_dist
B_node.set_dist(B_skill_dist)
C_skill_dist = DiscreteDistribution(C_node)
index = C_skill_dist.generate_index([], [])
C_skill_dist[index] = prior_skill_dist
C_node.set_dist(C_skill_dist)
# Match Prob Distribution
# P(T1vT2 | T1, T2)
# T1 | T2 | T1 | T2 | Tie
# 0 | 0 | 0.1 | 0.1 | 0.8
# 0 | 1 | 0.2 | 0.6 | 0.2
# 0 | 2 | 0.15 | 0.75 | 0.1
# 0 | 3 | 0.05 | 0.9 | 0.05
# 1 | 0 | 0.6 | 0.2 | 0.2
# 1 | 1 | 0.1 | 0.1 | 0.8
# 1 | 2 | 0.2 | 0.6 | 0.2
# 1 | 3 | 0.15 | 0.75 | 0.1
# 2 | 0 | 0.75 | 0.15 | 0.1
# 2 | 1 | 0.6 | 0.2 | 0.2
# 2 | 2 | 0.1 | 0.1 | 0.8
# 2 | 3 | 0.2 | 0.6 | 0.2
# 3 | 0 | 0.9 | 0.05 | 0.05
# 3 | 1 | 0.75 | 0.15 | 0.1
# 3 | 2 | 0.6 | 0.2 | 0.2
# 3 | 3 | 0.1 | 0.1 | 0.8
match_dist = zeros([A_node.size(), B_node.size(), AvB_node.size()], dtype=float32)
match_dist[0, 0,:] = [0.1, 0.1, 0.8]
match_dist[0, 1,:] = [0.2, 0.6, 0.2]
match_dist[0, 2,:] = [0.15, 0.75, 0.1]
match_dist[0, 3,:] = [0.05, 0.9, 0.05]
match_dist[1, 0,:] = [0.6, 0.2, 0.2]
match_dist[1, 1,:] = [0.1, 0.1, 0.8]
match_dist[1, 2,:] = [0.2, 0.6, 0.2]
match_dist[1, 3,:] = [0.15, 0.75, 0.1]
match_dist[2, 0,:] = [0.75, 0.15, 0.1]
match_dist[2, 1,:] = [0.6, 0.2, 0.2]
match_dist[2, 2,:] = [0.1, 0.1, 0.8]
match_dist[2, 3,:] = [0.2, 0.6, 0.2]
match_dist[3, 0,:] = [0.9, 0.05, 0.05]
match_dist[3, 1,:] = [0.75, 0.15, 0.1]
match_dist[3, 2,:] = [0.6, 0.2, 0.2]
match_dist[3, 3,:] = [0.1, 0.1, 0.8]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A_node, B_node, AvB_node],
table=match_dist)
AvB_node.set_dist(AvB_distribution)
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node],
table=match_dist)
BvC_node.set_dist(BvC_distribution)
CvA_distribution = ConditionalDiscreteDistribution(nodes=[C_node, A_node, CvA_node],
table=match_dist)
CvA_node.set_dist(CvA_distribution)
return BayesNet(nodes)
def get_test_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodes = []
# TODO: fill this out
# Create Nodes
A_node = BayesNode(0, 4, name="A")
B_node = BayesNode(1, 4, name="B")
C_node = BayesNode(2, 4, name="C")
D_node = BayesNode(3, 4, name="D")
E_node = BayesNode(4, 4, name="E")
AvB_node = BayesNode(5, 3, name="AvB")
BvC_node = BayesNode(6, 3, name="BvC")
CvD_node = BayesNode(7, 3, name="CvD")
DvE_node = BayesNode(8, 3, name="DvE")
EvA_node = BayesNode(9, 3, name="EvA")
# A skill
A_node.add_child(AvB_node)
A_node.add_child(EvA_node)
# B skill
B_node.add_child(AvB_node)
B_node.add_child(BvC_node)
# C skill
C_node.add_child(BvC_node)
C_node.add_child(CvD_node)
# D skill
B_node.add_child(CvD_node)
B_node.add_child(DvE_node)
# E skill
C_node.add_child(DvE_node)
C_node.add_child(EvA_node)
# AvB outcome
AvB_node.add_parent(A_node)
AvB_node.add_parent(B_node)
# BvC outcome
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
# CvD outcome
CvD_node.add_parent(C_node)
CvD_node.add_parent(D_node)
# DvE outcome
DvE_node.add_parent(D_node)
DvE_node.add_parent(E_node)
# EvA outcome
EvA_node.add_parent(E_node)
EvA_node.add_parent(A_node)
# Set Probability Distributions
# A, B, C team skill levels
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.15, 0.45, 0.3, 0.1]
A_node.set_dist(A_distribution)
# B_node.set_dist(A_distribution)
# C_node.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.15, 0.45, 0.3, 0.1]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([], [])
C_distribution[index] = [0.15, 0.45, 0.3, 0.1]
C_node.set_dist(C_distribution)
D_distribution = DiscreteDistribution(D_node)
index = D_distribution.generate_index([], [])
D_distribution[index] = [0.15, 0.45, 0.3, 0.1]
D_node.set_dist(D_distribution)
E_distribution = DiscreteDistribution(E_node)
index = E_distribution.generate_index([], [])
E_distribution[index] = [0.15, 0.45, 0.3, 0.1]
E_node.set_dist(E_distribution)
# AvB, BvC, CvA game outcomes
dist = zeros([A_node.size(), B_node.size(), AvB_node.size()], dtype=float32)
dist[0, 0, :] = [0.1, 0.1, 0.8]
dist[0, 1, :] = [0.2, 0.6, 0.2]
dist[0, 2, :] = [0.15, 0.75, 0.1]
dist[0, 3, :] = [0.05, 0.9, 0.05]
dist[1, 0, :] = [0.6, 0.2, 0.2]
dist[1, 1, :] = [0.1, 0.1, 0.8]
dist[1, 2, :] = [0.2, 0.6, 0.2]
dist[1, 3, :] = [0.15, 0.75, 0.1]
dist[2, 0, :] = [0.75, 0.15, 0.1]
dist[2, 1, :] = [0.6, 0.2, 0.2]
dist[2, 2, :] = [0.1, 0.1, 0.8]
dist[2, 3, :] = [0.2, 0.6, 0.2]
dist[3, 0, :] = [0.9, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.1]
dist[3, 2, :] = [0.6, 0.2, 0.2]
dist[3, 3, :] = [0.1, 0.1, 0.8]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
CvD_distribution = ConditionalDiscreteDistribution(nodes=[C_node, D_node, CvD_node], table=dist)
CvD_node.set_dist(CvD_distribution)
DvE_distribution = ConditionalDiscreteDistribution(nodes=[D_node, E_node, DvE_node], table=dist)
DvE_node.set_dist(DvE_distribution)
EvA_distribution = ConditionalDiscreteDistribution(nodes=[E_node, A_node, EvA_node], table=dist)
EvA_node.set_dist(EvA_distribution)
nodes = [A_node, B_node, C_node, D_node, E_node, AvB_node, BvC_node, CvD_node, DvE_node, EvA_node]
return BayesNet(nodes)
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodenames = ["A", "B", "C", "AvB", "BvC", "CvA"]
A_node = BayesNode(0, 4, name=nodenames[0])
B_node = BayesNode(1, 4, name=nodenames[1])
C_node = BayesNode(2, 4, name=nodenames[2])
AvB_node = BayesNode(3, 3, name=nodenames[3])
BvC_node = BayesNode(4, 3, name=nodenames[4])
CvA_node = BayesNode(5, 3, name=nodenames[5])
A_node.add_child(AvB_node)
A_node.add_child(CvA_node)
B_node.add_child(AvB_node)
B_node.add_child(BvC_node)
C_node.add_child(BvC_node)
C_node.add_child(CvA_node)
AvB_node.add_parent(A_node)
AvB_node.add_parent(B_node)
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
CvA_node.add_parent(C_node)
CvA_node.add_parent(A_node)
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.15, 0.45, 0.30, 0.10]
A_node.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.15, 0.45, 0.30, 0.10]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([], [])
C_distribution[index] = [0.15, 0.45, 0.30, 0.10]
C_node.set_dist(C_distribution)
dist = zeros([A_node.size(), B_node.size(),
AvB_node.size()],
dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
AvB_distribution = ConditionalDiscreteDistribution(
nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
dist = zeros([B_node.size(), C_node.size(),
BvC_node.size()],
dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
BvC_distribution = ConditionalDiscreteDistribution(
nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
dist = zeros([C_node.size(), A_node.size(),
CvA_node.size()],
dtype=float32)
dist[0, 0, :] = [0.10, 0.10, 0.80]
dist[0, 1, :] = [0.20, 0.60, 0.20]
dist[0, 2, :] = [0.15, 0.75, 0.10]
dist[0, 3, :] = [0.05, 0.90, 0.05]
dist[1, 0, :] = [0.60, 0.20, 0.20]
dist[1, 1, :] = [0.10, 0.10, 0.80]
dist[1, 2, :] = [0.20, 0.60, 0.20]
dist[1, 3, :] = [0.15, 0.75, 0.10]
dist[2, 0, :] = [0.75, 0.15, 0.10]
dist[2, 1, :] = [0.60, 0.20, 0.20]
dist[2, 2, :] = [0.10, 0.10, 0.80]
dist[2, 3, :] = [0.20, 0.60, 0.20]
dist[3, 0, :] = [0.90, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.10]
dist[3, 2, :] = [0.60, 0.20, 0.20]
dist[3, 3, :] = [0.10, 0.10, 0.80]
CvA_distribution = ConditionalDiscreteDistribution(
nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_distribution)
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
return BayesNet(nodes)
# TODO: fill this out
raise NotImplementedError
def make_exam_net():
"""Create a Bayes Net representation of the above power plant problem.
Name the nodes as "alarm","faulty alarm", "gauge","faulty gauge", "temperature".
"""
nodes = []
# TODO: finish this function
#create nodes ---> refer node.py
MA = BayesNode(0, 2, name="main alarm")
A1 = BayesNode(1, 2, name="alarm 2")
A2 = BayesNode(2, 2, name="alarm 1")
A3 = BayesNode(3, 2, name="alarm 3")
G = BayesNode(4, 2, name="ghost")
B = BayesNode(5, 2, name="burglar")
MA.add_parent(G)
MA.add_parent(B)
dist = zeros([G.size(), B.size(), MA.size()], dtype=float32)
dist[0,0,:] = [0.99, 0.01]
dist[0,1,:] = [0.36, 0.64]
dist[1,0,:] = [0.75, 0.25]
dist[1,1,:] = [0.02, 0.98]
MA_distribution = ConditionalDiscreteDistribution(nodes=[G, B, MA], table=dist)
MA.set_dist(MA_distribution)
G.add_child(MA)
G.add_child(A2)
G.add_child(A1)
G_distribution = DiscreteDistribution(G)
index = G_distribution.generate_index([],[])
G_distribution[index] = [0.6,0.4]
G.set_dist(G_distribution)
B.add_child(MA)
B.add_child(A2)
B.add_child(A3)
B_distribution = DiscreteDistribution(B)
index = B_distribution.generate_index([],[])
B_distribution[index] = [0.68,0.32]
B.set_dist(B_distribution)
A1.add_parent(G)
A1_distribution = DiscreteDistribution(A1)
dist = zeros([G.size(), A1.size()], dtype=float32) #Note the order of G_node, A_node
dist[0,:] = [0.91, 0.09]
dist[1,:] = [0.18, 0.82]
A1_distribution = ConditionalDiscreteDistribution(nodes=[G,A1], table=dist)
A1.set_dist(A1_distribution)
A2.add_parent(G)
A2.add_parent(B)
dist = zeros([G.size(), B.size(), A2.size()], dtype=float32)
dist[0,0,:] = [0.95, 0.05]
dist[0,1,:] = [0.37, 0.63]
dist[1,0,:] = [0.69, 0.31]
dist[1,1,:] = [0.13, 0.87]
A2_distribution = ConditionalDiscreteDistribution(nodes=[G, B, A2], table=dist)
A2.set_dist(A2_distribution)
A3.add_parent(B)
A3_distribution = DiscreteDistribution(A3)
dist = zeros([B.size(), A3.size()], dtype=float32) #Note the order of G_node, A_node
dist[0,:] = [0.28, 0.72]
dist[1,:] = [0.81, 0.19]
A3_distribution = ConditionalDiscreteDistribution(nodes=[B,A3], table=dist)
A3.set_dist(A3_distribution)
nodes = [MA,A1,A2,A3,G,B]
return BayesNet(nodes)
def get_game_network():
"""A Bayes Net representation
of the game problem."""
nodes = []
# TODO: fill this out
A_node = BayesNode(0, 4, name='A')
B_node = BayesNode(1, 4, name='B')
C_node = BayesNode(2, 4, name='C')
AvB_node = BayesNode(3, 3, name='A vs B')
BvC_node = BayesNode(4, 3, name='B vs C')
CvA_node = BayesNode(5, 3, name='C vs A')
nodes = []
A_node.add_child(AvB_node)
AvB_node.add_parent(A_node)
B_node.add_child(AvB_node)
AvB_node.add_parent(B_node)
B_node.add_child(BvC_node)
BvC_node.add_parent(B_node)
C_node.add_child(BvC_node)
BvC_node.add_parent(C_node)
C_node.add_child(CvA_node)
CvA_node.add_parent(C_node)
A_node.add_child(CvA_node)
CvA_node.add_parent(A_node)
nodes.append(A_node)
nodes.append(B_node)
nodes.append(C_node)
nodes.append(AvB_node)
nodes.append(BvC_node)
nodes.append(CvA_node)
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([],[])
A_distribution[index] = [0.15,0.45,0.3,0.1]
A_node.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([],[])
B_distribution[index] = [0.15,0.45,0.3,0.1]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([],[])
C_distribution[index] = [0.15,0.45,0.3,0.1]
C_node.set_dist(C_distribution)
dist = zeros([A_node.size(), B_node.size(), AvB_node.size()], dtype=float32)
# T2-T1=0
dist[0,0,:] = [0.10,0.10,0.80]
dist[1,1,:] = [0.10,0.10,0.80]
dist[2,2,:] = [0.10,0.10,0.80]
dist[3,3,:] = [0.10,0.10,0.80]
# T2-T1=1
dist[0,1,:] = [0.20,0.60,0.20]
dist[1,2,:] = [0.20,0.60,0.20]
dist[2,3,:] = [0.20,0.60,0.20]
dist[1,0,:] = [0.60,0.20,0.20]
dist[2,1,:] = [0.60,0.20,0.20]
dist[3,2,:] = [0.60,0.20,0.20]
# T2-T1=2
dist[0,2,:] = [0.15,0.75,0.10]
dist[1,3,:] = [0.15,0.75,0.10]
dist[2,0,:] = [0.75,0.15,0.10]
dist[3,1,:] = [0.75,0.15,0.10]
# T2-T1=3
dist[0,3,:] = [0.05,0.90,0.05]
dist[3,0,:] = [0.90,0.05,0.05]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
# P(BvC|B,C)
dist = zeros([B_node.size(), C_node.size(), BvC_node.size()], dtype=float32)
# T2-T1=0
dist[0,0,:] = [0.10,0.10,0.80]
dist[1,1,:] = [0.10,0.10,0.80]
dist[2,2,:] = [0.10,0.10,0.80]
dist[3,3,:] = [0.10,0.10,0.80]
# T2-T1=1
dist[0,1,:] = [0.20,0.60,0.20]
dist[1,2,:] = [0.20,0.60,0.20]
dist[2,3,:] = [0.20,0.60,0.20]
dist[1,0,:] = [0.60,0.20,0.20]
dist[2,1,:] = [0.60,0.20,0.20]
dist[3,2,:] = [0.60,0.20,0.20]
# T2-T1=2
dist[0,2,:] = [0.15,0.75,0.10]
dist[1,3,:] = [0.15,0.75,0.10]
dist[2,0,:] = [0.75,0.15,0.10]
dist[3,1,:] = [0.75,0.15,0.10]
# T2-T1=3
dist[0,3,:] = [0.05,0.90,0.05]
dist[3,0,:] = [0.90,0.05,0.05]
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
# P(CvA|C,A)
dist = zeros([C_node.size(), A_node.size(), CvA_node.size()], dtype=float32)
# T2-T1=0
dist[0,0,:] = [0.10,0.10,0.80]
dist[1,1,:] = [0.10,0.10,0.80]
dist[2,2,:] = [0.10,0.10,0.80]
dist[3,3,:] = [0.10,0.10,0.80]
# T2-T1=1
dist[0,1,:] = [0.20,0.60,0.20]
dist[1,2,:] = [0.20,0.60,0.20]
dist[2,3,:] = [0.20,0.60,0.20]
dist[1,0,:] = [0.60,0.20,0.20]
dist[2,1,:] = [0.60,0.20,0.20]
dist[3,2,:] = [0.60,0.20,0.20]
# T2-T1=2
dist[0,2,:] = [0.15,0.75,0.10]
dist[1,3,:] = [0.15,0.75,0.10]
dist[2,0,:] = [0.75,0.15,0.10]
dist[3,1,:] = [0.75,0.15,0.10]
# T2-T1=3
dist[0,3,:] = [0.05,0.90,0.05]
dist[3,0,:] = [0.90,0.05,0.05]
CvA_distribution = ConditionalDiscreteDistribution(nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_distribution)
return BayesNet(nodes)
def get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodes = []
# TODO: fill this out
#raise NotImplementedError
#Skill level Nodes: each has 0-3 four levels
A_node=BayesNode(0, 4, name='A')
B_node = BayesNode(1, 4, name='B')
C_node = BayesNode(2, 4, name='C')
# match nodes: each has win, lose or tie
AvB_node=BayesNode(3, 3, name='AvB')
BvC_node = BayesNode(4, 3, name='BvC')
CvA_node=BayesNode(5, 3, name='CvA')
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
#create network
A_node.add_child(AvB_node)
AvB_node.add_parent(A_node)
A_node.add_child(CvA_node)
CvA_node.add_parent(A_node)
B_node.add_child(AvB_node)
AvB_node.add_parent(B_node)
B_node.add_child(BvC_node)
BvC_node.add_parent(B_node)
C_node.add_child(CvA_node)
CvA_node.add_parent(C_node)
C_node.add_child(BvC_node)
BvC_node.add_parent(C_node)
#set Probability
# each team has 4 level of skills with probability 0.15, 0.45, 0.3 0.1
A_distribution = DiscreteDistribution(A_node)
index = A_distribution.generate_index([], [])
A_distribution[index] = [0.15, 0.45, 0.30, 0.10]
A_node.set_dist(A_distribution)
B_distribution = DiscreteDistribution(B_node)
index = B_distribution.generate_index([], [])
B_distribution[index] = [0.15, 0.45, 0.30, 0.10]
B_node.set_dist(B_distribution)
C_distribution = DiscreteDistribution(C_node)
index = C_distribution.generate_index([], [])
C_distribution[index] = [0.15, 0.45, 0.30, 0.10]
C_node.set_dist(C_distribution)
#Probability of matches
#AvB: given skill level A, B, P(AvB|A,B)
dist = zeros([A_node.size(), B_node.size(), AvB_node.size()], dtype=float32)
dist[0, 0, :] = [0.1, 0.1, 0.8]
dist[0, 1, :] = [0.2, 0.6, 0.2]
dist[0, 2, :] = [0.15, 0.75, 0.1]
dist[0, 3, :] = [0.05, 0.9, 0.05]
dist[1, 0, :] = [0.6, 0.2, 0.2]
dist[1, 1, :] = [0.1, 0.1, 0.8]
dist[1, 2, :] = [0.2, 0.6, 0.2]
dist[1, 3, :] = [0.15, 0.75, 0.1]
dist[2, 0, :] = [0.75, 0.15, 0.1]
dist[2, 1, :] = [0.6, 0.2, 0.2]
dist[2, 2, :] = [0.1, 0.1, 0.8]
dist[2, 3, :] = [0.2, 0.6, 0.2]
dist[3, 0, :] = [0.9, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.1]
dist[3, 2, :] = [0.6, 0.2, 0.2]
dist[3, 3, :] = [0.1, 0.1, 0.8]
AvB_distribution = ConditionalDiscreteDistribution(nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_distribution)
# BvC: given skill level B, C, P(BvC|B,C)
dist = zeros([B_node.size(), C_node.size(), BvC_node.size()], dtype=float32)
dist[0, 0, :] = [0.1, 0.1, 0.8]
dist[0, 1, :] = [0.2, 0.6, 0.2]
dist[0, 2, :] = [0.15, 0.75, 0.1]
dist[0, 3, :] = [0.05, 0.9, 0.05]
dist[1, 0, :] = [0.6, 0.2, 0.2]
dist[1, 1, :] = [0.1, 0.1, 0.8]
dist[1, 2, :] = [0.2, 0.6, 0.2]
dist[1, 3, :] = [0.15, 0.75, 0.1]
dist[2, 0, :] = [0.75, 0.15, 0.1]
dist[2, 1, :] = [0.6, 0.2, 0.2]
dist[2, 2, :] = [0.1, 0.1, 0.8]
dist[2, 3, :] = [0.2, 0.6, 0.2]
dist[3, 0, :] = [0.9, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.1]
dist[3, 2, :] = [0.6, 0.2, 0.2]
dist[3, 3, :] = [0.1, 0.1, 0.8]
BvC_distribution = ConditionalDiscreteDistribution(nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_distribution)
# CvA: given skill level C, A, P(CvA|C,A)
dist = zeros([C_node.size(), A_node.size(), CvA_node.size()], dtype=float32)
dist[0, 0, :] = [0.1, 0.1, 0.8]
dist[0, 1, :] = [0.2, 0.6, 0.2]
dist[0, 2, :] = [0.15, 0.75, 0.1]
dist[0, 3, :] = [0.05, 0.9, 0.05]
dist[1, 0, :] = [0.6, 0.2, 0.2]
dist[1, 1, :] = [0.1, 0.1, 0.8]
dist[1, 2, :] = [0.2, 0.6, 0.2]
dist[1, 3, :] = [0.15, 0.75, 0.1]
dist[2, 0, :] = [0.75, 0.15, 0.1]
dist[2, 1, :] = [0.6, 0.2, 0.2]
dist[2, 2, :] = [0.1, 0.1, 0.8]
dist[2, 3, :] = [0.2, 0.6, 0.2]
dist[3, 0, :] = [0.9, 0.05, 0.05]
dist[3, 1, :] = [0.75, 0.15, 0.1]
dist[3, 2, :] = [0.6, 0.2, 0.2]
dist[3, 3, :] = [0.1, 0.1, 0.8]
CvA_distribution = ConditionalDiscreteDistribution(nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_distribution)
return BayesNet(nodes)