def make_final_exam_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
# TODO: finish this function
N_node = BayesNode(0, 2, name = 'netflix')
E_node = BayesNode(1, 2, name = 'exercise')
A_node = BayesNode(2, 2, name = 'assignment')
C_node = BayesNode(3, 2, name = 'club')
S_node = BayesNode(4, 2, name = 'social')
G_node = BayesNode(5, 2, name='graduate')
nodes.extend([N_node, E_node, A_node, C_node, S_node, G_node])
A_node.add_parent(N_node)
N_node.add_child(A_node)
A_node.add_parent(E_node)
E_node.add_child(A_node)
S_node.add_parent(C_node)
C_node.add_child(S_node)
G_node.add_parent(A_node)
A_node.add_child(G_node)
G_node.add_parent(S_node)
S_node.add_child(G_node)
return BayesNet(nodes)
def make_cargo_flight_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
# TODO: finish this function
paint_node = BayesNode(0, 2, name = 'paint')
weather_node = BayesNode(1, 2, name = 'weather')
oring_node = BayesNode(2, 2, name = 'oring')
shuttle_node = BayesNode(3, 2, name = 'shuttle')
booster_node = BayesNode(4, 2, name = 'booster')
mission_node = BayesNode(5, 2, name='mission')
nodes.extend([paint_node, weather_node, oring_node, shuttle_node, booster_node, mission_node])
shuttle_node.add_parent(paint_node)
paint_node.add_child(shuttle_node)
shuttle_node.add_parent(weather_node)
weather_node.add_child(shuttle_node)
booster_node.add_parent(weather_node)
weather_node.add_child(booster_node)
booster_node.add_parent(oring_node)
oring_node.add_child(booster_node)
mission_node.add_parent(shuttle_node)
shuttle_node.add_child(mission_node)
mission_node.add_parent(booster_node)
booster_node.add_child(mission_node)
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
G_node = BayesNode(0, 2, name='gauge')
FG_node = BayesNode(1, 2, name='faulty gauge')
T_node = BayesNode(2, 2, name='temperature')
A_node = BayesNode(3, 2, name='alarm')
FA_node = BayesNode(4, 2, name='faulty alarm')
T_node.add_child(G_node)
T_node.add_child(FG_node)
G_node.add_parent(T_node)
FG_node.add_parent(T_node)
FG_node.add_child(G_node)
G_node.add_parent(FG_node)
G_node.add_child(A_node)
A_node.add_parent(G_node)
FA_node.add_child(A_node)
A_node.add_parent(FA_node)
nodes = [G_node, FG_node, T_node, FA_node, A_node]
# TODO: finish this function
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
A_node = BayesNode(0, 2, name='alarm')
Fa_node = BayesNode(1, 2, name='faulty alarm')
G_node = BayesNode(2, 2, name='gauge')
Fg_node = BayesNode(3, 2, name='faulty gauge')
T_node = BayesNode(4, 2, name='temperature')
# Faulty alarm and gauge reading affects the alarm
A_node.add_parent(Fa_node)
Fa_node.add_child(A_node)
A_node.add_parent(G_node)
G_node.add_child(A_node)
# Faulty gauge and temperature affects the gauge
G_node.add_parent(Fg_node)
Fg_node.add_child(G_node)
G_node.add_parent(T_node)
T_node.add_child(G_node)
# High temperature can cause faulty gauge
Fg_node.add_parent(T_node)
T_node.add_child(Fg_node)
nodes = [A_node, Fa_node, G_node, Fg_node, T_node]
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
A_node = BayesNode(0, 2, name = "alarm")
FA_node = BayesNode(1, 2, name = "faulty alarm")
G_node = BayesNode(2, 2, name = "gauge")
FG_node = BayesNode(3, 2, name = "faulty gauge")
T_node = BayesNode(4, 2, name = "temperature")
T_node.add_child(FG_node)
T_node.add_child(G_node)
FG_node.add_child(G_node)
G_node.add_child(A_node)
FA_node.add_child(A_node)
G_node.add_parent(T_node)
G_node.add_parent(FG_node)
FG_node.add_parent(T_node)
A_node.add_parent(G_node)
A_node.add_parent(FA_node)
nodes = [A_node, FA_node, G_node, FG_node, T_node]
return BayesNet(nodes)
raise NotImplementedError
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge",
"temperature". (for the tests to work.)
"""
nodenames = [
"temperature", "faulty gauge", "gauge", "faulty alarm", "alarm"
]
T_node = BayesNode(0, 2, name=nodenames[0])
Fg_node = BayesNode(1, 2, name=nodenames[1])
G_node = BayesNode(2, 2, name=nodenames[2])
Fa_node = BayesNode(3, 2, name=nodenames[3])
A_node = BayesNode(4, 2, name=nodenames[4])
T_node.add_child(Fg_node)
T_node.add_child(G_node)
Fg_node.add_parent(T_node)
Fg_node.add_child(G_node)
G_node.add_parent(Fg_node)
G_node.add_parent(T_node)
G_node.add_child(A_node)
Fa_node.add_child(A_node)
A_node.add_parent(G_node)
A_node.add_parent(Fa_node)
nodes = [T_node, Fg_node, G_node, Fa_node, A_node]
return BayesNet(nodes)
# TODO: finish this function
raise NotImplementedError
def make_power_plant_net():
#testing basic bayes net class implementation
numberOfNodes = 5
#name the nodes
alarm = 0
faulty_alarm = 1
gauge = 2
faulty_gauge = 3
temperature = 4
#create nodes ---> refer Node.py
alarm_node = BayesNode(0, 2, name="alarm")
faulty_alarm_node = BayesNode(1, 2, name="faulty alarm")
gauge_node = BayesNode(2, 2, name="gauge")
faulty_gauge_node = BayesNode(3, 2, name="faulty gauge")
temperature_node = BayesNode(4, 2, name="temperature")
temperature_node.add_child(faulty_gauge_node)
temperature_node.add_child(gauge_node)
faulty_alarm_node.add_child(alarm_node)
faulty_gauge_node.add_parent(temperature_node)
faulty_gauge_node.add_child(gauge_node)
gauge_node.add_parent(faulty_gauge_node)
gauge_node.add_parent(temperature_node)
gauge_node.add_child(alarm_node)
alarm_node.add_parent(faulty_alarm_node)
alarm_node.add_parent(gauge_node)
nodes = [alarm_node, faulty_alarm_node, gauge_node,faulty_gauge_node,temperature_node]
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of
the above power plant problem."""
nodes = []
A_node = BayesNode(0, 2, name='alarm')
FA_node = BayesNode(1, 2, name='faulty alarm')
G_node = BayesNode(2, 2, name='gauge')
FG_node = BayesNode(3, 2, name='faulty gauge')
T_node = BayesNode(4, 2, name='temperature')
T_node.add_child(FG_node)
FG_node.add_parent(T_node)
T_node.add_child(G_node)
G_node.add_parent(T_node)
FG_node.add_child(G_node)
G_node.add_parent(FG_node)
G_node.add_child(A_node)
A_node.add_parent(G_node)
FA_node.add_child(A_node)
A_node.add_parent(FA_node)
nodes.append(A_node)
nodes.append(FA_node)
nodes.append(G_node)
nodes.append(FG_node)
nodes.append(T_node)
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
# TODO: finish this function
#raise NotImplementedError
# create nodes
A_node = BayesNode(0, 2, name='alarm')
F_A_node= BayesNode(1, 2, name='faulty alarm')
G_node=BayesNode(2, 2, name='gauge')
F_G_node = BayesNode(3, 2, name='faulty gauge')
T_node = BayesNode(4, 2, name='temperature')
#connect nodes
T_node.add_child(G_node)
G_node.add_parent(T_node)
T_node.add_child(F_G_node)
F_G_node.add_parent(T_node)
F_G_node.add_child(G_node)
G_node.add_parent(F_G_node)
G_node.add_child(A_node)
A_node.add_parent(G_node)
F_A_node.add_child(A_node)
A_node.add_parent(F_A_node)
nodes=[A_node,F_A_node,G_node,F_G_node,T_node]
return BayesNet(nodes)
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 get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
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)
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(A_node)
CvA_node.add_parent(C_node)
A_node = set_skill_distribution(A_node)
B_node = set_skill_distribution(B_node)
C_node = set_skill_distribution(C_node)
AvB_node = set_skill_diff_distribution(AvB_node,
[A_node, B_node, AvB_node])
BvC_node = set_skill_diff_distribution(BvC_node,
[B_node, C_node, BvC_node])
CvA_node = set_skill_diff_distribution(CvA_node,
[C_node, A_node, CvA_node])
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
return BayesNet(nodes)
def make_power_plant_net():
# intial setup
nodes = []
numberOfNodes = 5
temperature = 0
gauge = 1
faulty_gauge = 2
alarm = 3
faulty_alarm = 4
# create the nodes
T_node = BayesNode(0, 2, name='temperature')
G_node = BayesNode(1, 2, name='gauge')
F_G_node = BayesNode(2, 2, name='faulty gauge')
A_node = BayesNode(3, 2, name='alarm')
F_A_node = BayesNode(4, 2, name='faulty alarm')
# temperature
T_node.add_child(G_node)
T_node.add_child(F_G_node)
# faulty gauge
F_G_node.add_child(G_node)
F_G_node.add_parent(T_node)
# gauge
G_node.add_parent(T_node)
G_node.add_parent(F_G_node)
G_node.add_child(A_node)
# faulty alarm
F_A_node.add_child(A_node)
# alarm
A_node.add_parent(G_node)
A_node.add_parent(F_A_node)
# add the nodes for setting up example network
nodes = [A_node, F_A_node, G_node, F_G_node, T_node]
return BayesNet(nodes)
def make_game_net():
nodes = []
# TODO: fill this out
# raise NotImplementedError
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
A_node.add_child(AvB_node)
A_node.add_child(CvA_node)
# B
B_node.add_child(BvC_node)
B_node.add_child(AvB_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)
# Append nodes to node list
nodes.append(A_node)
nodes.append(B_node)
nodes.append(C_node)
nodes.append(AvB_node)
nodes.append(BvC_node)
nodes.append(CvA_node)
return BayesNet(nodes)
def make_power_plant_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
A = BayesNode(0, 2, name="alarm")
F_A = BayesNode(1, 2, name="faulty alarm")
G_node = BayesNode(2, 2, name="gauge")
F_G_node = BayesNode(3, 2, name="faulty gauge")
T_node = BayesNode(4, 2, name="temperature")
#alarm
A.add_parent(G_node)
A.add_parent(F_A)
#faulty alarm
F_A.add_child(A)
#gauge
G_node.add_parent(T_node)
G_node.add_parent(F_G_node)
G_node.add_child(A)
#faulty gauge
F_G_node.add_parent(T_node)
F_G_node.add_child(G_node)
#temperature
T_node.add_child(G_node)
T_node.add_child(F_G_node)
nodes = [A, F_A, G_node, F_G_node, T_node]
return BayesNet(nodes)
def make_power_plant_net():
"""Create a Bayes Net representation of the above power plant problem.
Use the following as the name attribute: "alarm","faulty alarm", "gauge","faulty gauge", "temperature". (for the tests to work.)
"""
nodes = []
# TODO: finish this function
A_Node = BayesNode(0, 2, name="alarm")
FA_Node = BayesNode(2, 2, name="faulty alarm")
G_Node = BayesNode(3, 2, name="gauge")
FG_Node = BayesNode(4, 2, name="faulty gauge")
T_Node = BayesNode(5, 2, name="temperature")
# T -> FG
T_Node.add_child(FG_Node)
FG_Node.add_parent(T_Node)
# FG -> G
FG_Node.add_child(G_Node)
G_Node.add_parent(FG_Node)
# T -> G
T_Node.add_child(G_Node)
G_Node.add_parent(T_Node)
# G -> A
G_Node.add_child(A_Node)
A_Node.add_parent(G_Node)
# FA -> A
FA_Node.add_child(A_Node)
A_Node.add_parent(FA_Node)
nodes.append(A_Node)
nodes.append(FA_Node)
nodes.append(G_Node)
nodes.append(FG_Node)
nodes.append(T_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". """
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')
# Skill level of A and B affects AvB
AvB_node.add_parent(A_node)
AvB_node.add_parent(B_node)
A_node.add_child(AvB_node)
B_node.add_child(AvB_node)
# Skill level of B and C affects BvC
BvC_node.add_parent(B_node)
BvC_node.add_parent(C_node)
B_node.add_child(BvC_node)
C_node.add_child(BvC_node)
# Skill level of C and A affects BvC
CvA_node.add_parent(A_node)
CvA_node.add_parent(C_node)
A_node.add_child(CvA_node)
C_node.add_child(CvA_node)
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_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_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)
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_dist = ConditionalDiscreteDistribution(
nodes=[A_node, B_node, AvB_node], table=dist)
AvB_node.set_dist(AvB_dist)
# The same distribution AvB applies for BvC and CvA
BvC_dist = ConditionalDiscreteDistribution(
nodes=[B_node, C_node, BvC_node], table=dist)
BvC_node.set_dist(BvC_dist)
CvA_dist = ConditionalDiscreteDistribution(
nodes=[C_node, A_node, CvA_node], table=dist)
CvA_node.set_dist(CvA_dist)
nodes = [A_node, B_node, C_node, AvB_node, BvC_node, CvA_node]
return BayesNet(nodes)
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 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():
"""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_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". """
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 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 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.
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 get_game_network():
"""Create a Bayes Net representation of the game problem.
Name the nodes as "A","B","C","AvB","BvC" and "CvA". """
nodes = []
A = BayesNode(0, 4, name="A")
B = BayesNode(1, 4, name="B")
C = BayesNode(2, 4, name="C")
AvB = BayesNode(3, 3, name="AvB")
BvC = BayesNode(4, 3, name="BvC")
CvA = BayesNode(5, 3, name="CvA")
# A -> AvB / A -> CvA
A.add_child(AvB)
A.add_child(CvA)
AvB.add_parent(A)
CvA.add_parent(A)
# B -> AvB / B -> BvC
B.add_child(AvB)
B.add_child(BvC)
AvB.add_parent(B)
BvC.add_parent(B)
# C -> BvC / C -> CvA
C.add_child(BvC)
C.add_child(CvA)
BvC.add_parent(C)
CvA.add_parent(C)
N = [A, B, C]
for n in N:
dist = DiscreteDistribution(n)
idx = dist.generate_index([], [])
dist[idx] = [0.15, 0.45, 0.3, 0.1]
n.set_dist(dist)
N = [AvB, BvC, CvA]
NN = [[A, B], [B, C], [C, A]]
c = 0
for n in N:
dist = zeros([NN[c][0].size(), NN[c][1].size(),
n.size()],
dtype=float32)
for i in range(0, 4):
for j in range(0, 4):
if ((j - i) == 0):
dist[i, j, :] = [0.1, 0.1, 0.8]
elif ((j - i) == 1):
dist[i, j, :] = [0.2, 0.6, 0.2]
elif ((j - i) == 2):
dist[i, j, :] = [0.15, 0.75, 0.1]
elif ((j - i) == 3):
dist[i, j, :] = [0.05, 0.9, 0.05]
elif ((j - i) == -1):
dist[i, j, :] = [0.6, 0.2, 0.2]
elif ((j - i) == -2):
dist[i, j, :] = [0.75, 0.15, 0.1]
elif ((j - i) == -3):
dist[i, j, :] = [0.9, 0.05, 0.05]
tmp = ConditionalDiscreteDistribution(nodes=[NN[c][0], NN[c][1], n],
table=dist)
n.set_dist(tmp)
c += 1
nodes = [A, B, C, AvB, BvC, CvA]
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 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():
"""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)