def main():
pa_nd = BayesNode(0, "pa_nd", size=4)
pa_nd.state_names = [str(k) for k in range(4)]
p_sh = PhaseShifter(1, "pshifter", pa_nd, 30, occ_nums=True)
print("pa_nd state names: ", pa_nd.state_names)
print("phase shifter state names: ", p_sh.state_names)
print(p_sh.potential)
print(p_sh.potential.get_total_probs())
def main():
pa_nd = BayesNode(0, "pa_nd", size=2)
pa_nd.state_names = ['0', '1']
thetas_degs = [20, 30, 40, 60]
qr = QubitRot(1, "rot", pa_nd, thetas_degs)
print("pa_nd state names: ", pa_nd.state_names)
print("qubit rot state names: ", qr.state_names)
print(qr.potential)
print(qr.potential.get_total_probs())
def read_bif(path, is_quantum):
"""
Reads a bif file using our stand-alone class BifTool and returns a
BayesNet. bif and dot files complement each other. bif: graphical
info No, pot info Yes. dot: graphical info Yes, pot info No. By pots
I mean potentials, the transition matrices of the nodes. (aka CPTs,
etc.)
Parameters
----------
path : str
is_quantum : bool
Returns
-------
BayesNet
"""
bt = BifTool(is_quantum)
bt.read_bif(path)
nodes = set()
name_to_nd = {}
for k, nd_name in enumerate(bt.nd_sizes.keys()):
node = BayesNode(k, nd_name)
node.state_names = bt.states[nd_name]
node.size = len(node.state_names)
node.forget_all_evidence()
nodes |= {node}
name_to_nd[nd_name] = node
for nd_name, pa_name_list in bt.parents.items():
node = name_to_nd[nd_name]
for pa_name in pa_name_list:
pa = name_to_nd[pa_name]
node.add_parent(pa)
for nd_name, parent_names in bt.parents.items():
node = name_to_nd[nd_name]
num_pa = len(parent_names)
parents = [name_to_nd[pa_name] for pa_name in parent_names]
if num_pa == 0:
node.potential = DiscreteUniPot(is_quantum, node)
else:
node.potential = DiscreteCondPot(
is_quantum, parents + [node])
node.potential.pot_arr = bt.pot_arrays[nd_name]
return BayesNet(nodes)
def __deepcopy__(self, memo):
"""
Creates deep copy of self.
Parameters
----------
memo : dict
Returns
-------
BayesNet
"""
nd_to_new_nd = {}
for nd in self.nodes:
nd_to_new_nd[nd] = BayesNode(nd.id_num, nd.name)
for nd, new_nd in nd_to_new_nd.items():
new_nd.neighbors = set([nd_to_new_nd[nd1] for nd1 in nd.neighbors])
new_nd.topo_index = nd.topo_index
new_nd.visited = nd.visited
new_nd.children = set([nd_to_new_nd[nd1] for nd1 in nd.children])
new_nd.parents = set([nd_to_new_nd[nd1] for nd1 in nd.parents])
# not in Dag:
new_nd.active_states = [x for x in nd.active_states]
new_pot_arr = cp.deepcopy(nd.potential.pot_arr)
new_ord_nodes = [
nd_to_new_nd[nd1] for nd1 in nd.potential.ord_nodes
]
new_nd.potential = Potential(nd.potential.is_quantum,
ord_nodes=new_ord_nodes,
pot_arr=new_pot_arr)
new_nd.size = nd.size
new_nd.state_names = [x for x in nd.state_names]
return BayesNet(set(nd_to_new_nd.values()))
def main():
theta_degs = 35
max_n_sum = 3
in_nd = BayesNode(0, "parent1", size=6)
in_nd.set_state_names_to_product([range(2), range(3)], trim=False)
pr = PolarizationRot(1, "pol_rot", in_nd, theta_degs, max_n_sum)
print("in_nd state names: ", in_nd.state_names)
print("pol rot state names: ", pr.state_names)
print(pr.potential)
print("full dict of total probs: ", pr.potential.get_total_probs())
print("brief dict of total probs: ",
pr.potential.get_total_probs(brief=True))
def main():
pa_nd = BayesNode(0, "pa_nd", size=24)
# has_commas = True
has_commas = False
pa_nd.set_state_names_to_product(
[range(2), range(3), range(4)], trim=not has_commas)
marg = Marginalizer(1,
"marg",
False,
pa_nd,
projected_axis=1,
has_commas=has_commas)
print("pa state names: ", pa_nd.state_names)
print("marg state names: ", marg.state_names)
print(marg.potential)
print(marg.potential.get_total_probs())
def main():
a = BayesNode(0, name="a")
b = BayesNode(1, name="b")
c = BayesNode(2, name="c")
a.add_neighbor(b)
a.add_neighbor(c)
assert Star.get_missing_edges_of_node(a) == [{b, c}]
def main():
pa1 = BayesNode(0, "parent1", size=2)
pa2 = BayesNode(1, "parent2", size=2)
pa1.state_names = ['0', '1']
pa2.state_names = ['0', '1']
cn = CNot(2, "a_cnot", False, pa1, pa2, True, True)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("cnot state names: ", cn.state_names)
print(cn.potential)
print(cn.potential.get_total_probs())
def new_from_nx_graph(nx_graph):
"""
Returns a BayesNet constructed from nx_graph.
Parameters
----------
nx_graph : networkx DiGraph
Returns
-------
BayesNet
"""
new_g = BayesNet(set())
for k, name in enumerate(nx_graph.nodes()):
new_g.add_nodes({BayesNode(k, name=name)})
node_list = list(new_g.nodes)
for nd in node_list:
for ch_name in nx_graph.successors(nd.name):
nd.add_child(new_g.get_node_named(ch_name))
return new_g
def build_bnet():
"""
Builds CBnet called WetGrass with diamond shape
Cloudy
/ \
Rain Sprinkler
\ /
WetGrass
All arrows pointing down
"""
cl = BayesNode(0, name="Cloudy")
sp = BayesNode(1, name="Sprinkler")
ra = BayesNode(2, name="Rain")
we = BayesNode(3, name="WetGrass")
we.add_parent(sp)
we.add_parent(ra)
sp.add_parent(cl)
ra.add_parent(cl)
nodes = {cl, ra, sp, we}
cl.potential = DiscreteUniPot(False, cl) # P(a)
sp.potential = DiscreteCondPot(False, [cl, sp]) # P(b| a)
ra.potential = DiscreteCondPot(False, [cl, ra])
we.potential = DiscreteCondPot(False, [sp, ra, we])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
cl.potential.pot_arr[:] = [.5, .5]
ra.potential.pot_arr[1, :] = [.5, .5]
ra.potential.pot_arr[0, :] = [.4, .6]
sp.potential.pot_arr[1, :] = [.7, .3]
sp.potential.pot_arr[0, :] = [.2, .8]
we.potential.pot_arr[1, 1, :] = [.01, .99]
we.potential.pot_arr[1, 0, :] = [.01, .99]
we.potential.pot_arr[0, 1, :] = [.01, .99]
we.potential.pot_arr[0, 0, :] = [.99, .01]
return BayesNet(nodes)
self.occ_nums = occ_nums
BayesNode.__init__(self, id_num, name, size=pa_nd.size)
self.add_parent(pa_nd)
self.state_names = pa_nd.state_names
pot = DiscreteCondPot(True, [pa_nd, self], bias=0)
self.potential = pot
theta_rads = theta_degs * math.pi / 180
for k in range(pa_nd.size):
if not occ_nums:
phase = theta_rads
else:
num = int(pa_nd.state_names[k])
phase = num * theta_rads
self.potential[k, k] = cmath.exp(1j * phase)
if __name__ == "__main__":
pa_nd = BayesNode(0, "pa_nd", size=4)
pa_nd.state_names = [str(k) for k in range(4)]
p_sh = PhaseShifter(1, "pshifter", pa_nd, 30, occ_nums=True)
print("pa_nd state names: ", pa_nd.state_names)
print("phase shifter state names: ", p_sh.state_names)
print(p_sh.potential)
print(p_sh.potential.get_total_probs())
def build_bnet():
"""
Builds simple 7 node binary tree
a0
/ \
b0 b1
/ \ / \
c0 c1,c2 c3
All arrows pointing down
"""
a0 = BayesNode(0, name="a0")
b0 = BayesNode(1, name="b0")
b1 = BayesNode(2, name="b1")
c0 = BayesNode(3, name="c0")
c1 = BayesNode(4, name="c1")
c2 = BayesNode(5, name="c2")
c3 = BayesNode(6, name="c3")
a0.add_children([b0, b1])
b0.add_children([c0, c1])
b1.add_children([c2, c3])
nodes = {a0, b0, b1, c0, c1, c2, c3}
a0.potential = DiscreteUniPot(False, a0) # P(a)
b0.potential = DiscreteCondPot(False, [a0, b0]) # P(b| a)
b1.potential = DiscreteCondPot(False, [a0, b1])
c0.potential = DiscreteCondPot(False, [b0, c0])
c1.potential = DiscreteCondPot(False, [b0, c1])
c2.potential = DiscreteCondPot(False, [b1, c2])
c3.potential = DiscreteCondPot(False, [b1, c3])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
a0.potential.pot_arr = np.array([.7, .3])
b0.potential.pot_arr = np.array([[.5, .5], [.4, .6]])
b1.potential.pot_arr = np.array([[.2, .8], [.4, .6]])
c0.potential.pot_arr = np.array([[.3, .7], [.4, .6]])
c1.potential.pot_arr = np.array([[.5, .5], [.9, .1]])
c2.potential.pot_arr = np.array([[.8, .2], [.4, .6]])
c3.potential.pot_arr = np.array([[.5, .5], [.6, .4]])
return BayesNet(nodes)
"""
if self.flipped_by_0:
x = 1 - pa1_st
else: # flipped by 1
x = pa1_st
if self.pa1_is_control:
bit0 = pa1_st
bit1 = (x + pa2_st) // 2
else: # pa2 is control
bit0 = (x + pa2_st) // 2
bit1 = pa2_st
foc_st = bit0 + 2 * bit1
return foc_st
if __name__ == "__main__":
pa1 = BayesNode(0, "parent1", size=2)
pa2 = BayesNode(1, "parent2", size=2)
pa1.state_names = ['0', '1']
pa2.state_names = ['0', '1']
cn = CNot(2, "a_cnot", False, pa1, pa2, True, True)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("cnot state names: ", cn.state_names)
print(cn.potential)
print(cn.potential.get_total_probs())
mx : list[int]
my : list[int]
Returns
-------
int
"""
return self.fill_trans_mat_and_st_names_of_nd(
mx, my, dry_run=True)
if __name__ == "__main__":
theta_degs = 35
max_n_sum = 3
in_nd = BayesNode(0, "parent1", size=6)
in_nd.set_state_names_to_product(
[range(2), range(3)], trim=False)
pr = PolarizationRot(1, "pol_rot", in_nd, theta_degs, max_n_sum)
print("in_nd state names: ", in_nd.state_names)
print("pol rot state names: ", pr.state_names)
print(pr.potential)
print("full dict of total probs: ",
pr.potential.get_total_probs())
print("brief dict of total probs: ",
pr.potential.get_total_probs(brief=True))
def build_bnet():
"""
Builds CBnet in accompanying gif : bnet_HuangDarwiche.gif
From "Inference Belief Networks: A Procedural Guide", by C.Huang and
A. Darwiche
"""
a_node = BayesNode(0, name="A")
b_node = BayesNode(1, name="B")
c_node = BayesNode(2, name="C")
d_node = BayesNode(3, name="D")
e_node = BayesNode(4, name="E")
f_node = BayesNode(5, name="F")
g_node = BayesNode(6, name="G")
h_node = BayesNode(7, name="H")
b_node.add_parent(a_node)
c_node.add_parent(a_node)
d_node.add_parent(b_node)
e_node.add_parent(c_node)
f_node.add_parent(d_node)
f_node.add_parent(e_node)
g_node.add_parent(c_node)
h_node.add_parent(e_node)
h_node.add_parent(g_node)
nodes = {
a_node, b_node, c_node, d_node, e_node, f_node, g_node, h_node
}
a_node.potential = DiscreteUniPot(False, a_node) # P(a)
b_node.potential = DiscreteCondPot(False, [a_node, b_node]) # P(b| a)
c_node.potential = DiscreteCondPot(False, [a_node, c_node])
d_node.potential = DiscreteCondPot(False, [b_node, d_node])
e_node.potential = DiscreteCondPot(False, [c_node, e_node])
# P(f|d, e)
f_node.potential = DiscreteCondPot(False, [d_node, e_node, f_node])
g_node.potential = DiscreteCondPot(False, [c_node, g_node])
h_node.potential = DiscreteCondPot(False, [e_node, g_node, h_node])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
a_node.potential.pot_arr[:] = [.5, .5]
b_node.potential.pot_arr[1, :] = [.5, .5]
b_node.potential.pot_arr[0, :] = [.4, .6]
c_node.potential.pot_arr[1, :] = [.7, .3]
c_node.potential.pot_arr[0, :] = [.2, .8]
d_node.potential.pot_arr[1, :] = [.9, .1]
d_node.potential.pot_arr[0, :] = [.5, .5]
e_node.potential.pot_arr[1, :] = [.3, .7]
e_node.potential.pot_arr[0, :] = [.6, .4]
f_node.potential.pot_arr[1, 1, :] = [.01, .99]
f_node.potential.pot_arr[1, 0, :] = [.01, .99]
f_node.potential.pot_arr[0, 1, :] = [.01, .99]
f_node.potential.pot_arr[0, 0, :] = [.99, .01]
g_node.potential.pot_arr[1, :] = [.8, .2]
g_node.potential.pot_arr[0, :] = [.1, .9]
h_node.potential.pot_arr[1, 1, :] = [.05, .95]
h_node.potential.pot_arr[1, 0, :] = [.95, .05]
h_node.potential.pot_arr[0, 1, :] = [.95, .05]
h_node.potential.pot_arr[0, 0, :] = [.95, .05]
return BayesNet(nodes)
def main():
num_of_comps = 1
tau_mag = .5
tau_degs = 35
rho_degs = 45
if num_of_comps == 1:
size1 = 3
size2 = 4
max_n_sum = 5
pa1 = BayesNode(0, "parent1", size=size1)
pa2 = BayesNode(1, "parent2", size=size2)
pa1.state_names = [str(k) for k in range(size1)]
pa2.state_names = [str(k) for k in range(size2)]
bs = BeamSplitter(2, "a_bs", pa1, pa2,
tau_mag, tau_degs, rho_degs, num_of_comps, max_n_sum)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("bs state names: ", bs.state_names)
print(bs.potential)
print("full dict of total probs: ",
bs.potential.get_total_probs())
print("brief dict of total probs: ",
bs.potential.get_total_probs(brief=True))
elif num_of_comps == 2:
size1 = 6
size2 = 8
max_n_sum = 7
pa1 = BayesNode(0, "parent1", size=size1)
pa2 = BayesNode(1, "parent2", size=size2)
pa1.set_state_names_to_product(
[range(2), range(3)], trim=False)
pa2.set_state_names_to_product(
[range(2), range(4)], trim=False)
bs = BeamSplitter(2, "a_bs", pa1, pa2,
tau_mag, tau_degs, rho_degs, num_of_comps, max_n_sum)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("bs state names: ", bs.state_names)
print(bs.potential)
print("full dict of total probs: ",
bs.potential.get_total_probs())
print("brief dict of total probs: ",
bs.potential.get_total_probs(brief=True))
def build_bnet():
"""
Builds simple 7 node binary tree
a0
/ \
b0 b1
/ \ / \
c0 c1,c2 c3
All arrows pointing down
"""
a0 = BayesNode(0, name="a0")
b0 = BayesNode(1, name="b0")
b1 = BayesNode(2, name="b1")
c0 = BayesNode(3, name="c0")
c1 = BayesNode(4, name="c1")
c2 = BayesNode(5, name="c2")
c3 = BayesNode(6, name="c3")
a0.add_children([b0, b1])
b0.add_children([c0, c1])
b1.add_children([c2, c3])
nodes = {a0, b0, b1, c0, c1, c2, c3}
a0.potential = DiscreteUniPot(False, a0) # P(a)
b0.potential = DiscreteCondPot(False, [a0, b0]) # P(b| a)
b1.potential = DiscreteCondPot(False, [a0, b1])
c0.potential = DiscreteCondPot(False, [b0, c0])
c1.potential = DiscreteCondPot(False, [b0, c1])
c2.potential = DiscreteCondPot(False, [b1, c2])
c3.potential = DiscreteCondPot(False, [b1, c3])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
a0.potential.pot_arr = np.array([.7, .3])
b0.potential.pot_arr = np.array([[.5, .5], [.4, .6]])
b1.potential.pot_arr = np.array([[.2, .8], [.4, .6]])
c0.potential.pot_arr = np.array([[.3, .7], [.4, .6]])
c1.potential.pot_arr = np.array([[.5, .5], [.9, .1]])
c2.potential.pot_arr = np.array([[.8, .2], [.4, .6]])
c3.potential.pot_arr = np.array([[.5, .5], [.6, .4]])
return BayesNet(nodes)
self.theta_degs = theta_degs
self.occ_nums = occ_nums
BayesNode.__init__(self, id_num, name, size=pa_nd.size)
self.add_parent(pa_nd)
self.state_names = pa_nd.state_names
pot = DiscreteCondPot(True, [pa_nd, self], bias=0)
self.potential = pot
theta_rads = theta_degs*math.pi/180
for k in range(pa_nd.size):
if not occ_nums:
phase = theta_rads
else:
num = int(pa_nd.state_names[k])
phase = num*theta_rads
self.potential[k, k] = cmath.exp(1j*phase)
if __name__ == "__main__":
pa_nd = BayesNode(0, "pa_nd", size=4)
pa_nd.state_names = [str(k) for k in range(4)]
p_sh = PhaseShifter(1, "pshifter", pa_nd, 30, occ_nums=True)
print("pa_nd state names: ", pa_nd.state_names)
print("phase shifter state names: ", p_sh.state_names)
print(p_sh.potential)
print(p_sh.potential.get_total_probs())
BayesNode.__init__(self, id_num, name, size=size)
self.add_parent(pa_nd)
self.state_names = non_rep_name_list
pot = DiscreteCondPot(is_quantum, [pa_nd, self], bias=0)
self.potential = pot
for k, name in enumerate(self.state_names):
for r, pa_name in enumerate(rep_name_list):
if name == pa_name:
# remember, focus node is last topo_index
self.potential[r, k] = 1
if __name__ == "__main__":
pa_nd = BayesNode(0, "pa_nd", size=24)
# has_commas = True
has_commas = False
pa_nd.set_state_names_to_product(
[range(2), range(3), range(4)], trim=not has_commas)
marg = Marginalizer(1, "marg",
False, pa_nd, projected_axis=1, has_commas=has_commas)
print("pa state names: ", pa_nd.state_names)
print("marg state names: ", marg.state_names)
print(marg.potential)
print(marg.potential.get_total_probs())
"""
return str([node.name for node in self.ord_nodes]) \
+ "\n" + str(self.pot_arr)
from nodes.BayesNode import *
if __name__ == "__main__":
with np.errstate(all='ignore'):
x = np.array([2, 0]) / np.array([1, 0])
x[x == np.inf] = 0
x = np.nan_to_num(x)
print("[2, 0]/[1, 0]=", x)
# define some nodes
a_node = BayesNode(0, name="A")
b_node = BayesNode(1, name="B")
c_node = BayesNode(2, name="C")
d_node = BayesNode(3, name="D")
e_node = BayesNode(4, name="E")
f_node = BayesNode(5, name="F")
g_node = BayesNode(6, name="G")
print("\ndefine some pots")
# Make sure if entries of numpy array are integers, to specify
# dtype=float64 or numpy will cast them as integers and this will lead
# to type casting errors
ar_ab = np.arange(4, dtype=np.float64).reshape(2, 2)
pot_ab = Potential(False, [a_node, b_node], ar_ab)
"""
if self.flipped_by_0:
x = 1 - pa1_st
else: # flipped by 1
x = pa1_st
if self.pa1_is_control:
bit0 = pa1_st
bit1 = (x + pa2_st)//2
else: # pa2 is control
bit0 = (x + pa2_st)//2
bit1 = pa2_st
foc_st = bit0 + 2*bit1
return foc_st
if __name__ == "__main__":
pa1 = BayesNode(0, "parent1", size=2)
pa2 = BayesNode(1, "parent2", size=2)
pa1.state_names = ['0', '1']
pa2.state_names = ['0', '1']
cn = CNot(2, "a_cnot",
False, pa1, pa2, True, True)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("cnot state names: ", cn.state_names)
print(cn.potential)
print(cn.potential.get_total_probs())
return self.fill_trans_mat_and_st_names_of_nd(
m1x, m2x, m1y, m2y, dry_run=True)
if __name__ == "__main__":
num_of_comps = 1
tau_mag = .5
tau_degs = 35
rho_degs = 45
if num_of_comps == 1:
size1 = 3
size2 = 4
max_n_sum = 5
pa1 = BayesNode(0, "parent1", size=size1)
pa2 = BayesNode(1, "parent2", size=size2)
pa1.state_names = [str(k) for k in range(size1)]
pa2.state_names = [str(k) for k in range(size2)]
bs = BeamSplitter(2, "a_bs", pa1, pa2,
tau_mag, tau_degs, rho_degs, num_of_comps, max_n_sum)
print("pa1 state names: ", pa1.state_names)
print("pa2 state names: ", pa2.state_names)
print("bs state names: ", bs.state_names)
print(bs.potential)
print("full dict of total probs: ",
bs.potential.get_total_probs())
print("brief dict of total probs: ",
Parameters
----------
nd : BayesNode
Returns
-------
list[set[BayesNode]]
"""
edges = list()
neig_list = list(nd.neighbors)
for k in range(len(neig_list)):
n1 = neig_list[k]
for n2 in neig_list[k + 1:]:
if not n2.has_neighbor(n1):
edges.append({n1, n2})
return edges
from nodes.Node import *
if __name__ == "__main__":
a = BayesNode(0, name="a")
b = BayesNode(1, name="b")
c = BayesNode(2, name="c")
a.add_neighbor(b)
a.add_neighbor(c)
assert (Star.get_missing_edges_of_node(a) == [{b, c}])
def main():
with np.errstate(all='ignore'):
x = np.array([2, 0 + 0j]) / np.array([1, 0])
x[~np.isfinite(x)] = 0
print("[2, 0+0j]/[1, 0]=", x)
with np.errstate(all='ignore'):
x = np.array([2, 5]) / np.array([1, 0])
x[~np.isfinite(x)] = 0
print("[2, 5]/[1, 0]=", x)
# define some nodes
a_node = BayesNode(0, name="A")
b_node = BayesNode(1, name="B")
c_node = BayesNode(2, name="C")
d_node = BayesNode(3, name="D")
e_node = BayesNode(4, name="E")
f_node = BayesNode(5, name="F")
g_node = BayesNode(6, name="G")
print("\n-----------------define some pots")
# Make sure if entries of numpy array are integers, to specify
# dtype=float64 or numpy will cast them as integers and this will lead
# to type casting errors
ar_ab = np.arange(4, dtype=np.float64).reshape(2, 2)
pot_ab = Potential(False, [a_node, b_node], ar_ab)
print("pot_ab:", pot_ab)
ar_ab2 = np.arange(0, 40, 10, dtype=np.float64).reshape(2, 2)
pot_ab2 = Potential(False, [a_node, b_node], ar_ab2)
print("pot_ab2:", pot_ab2)
ar_bc = np.arange(0, 40, 10, dtype=np.float64).reshape(2, 2)
pot_bc = Potential(False, [b_node, c_node], ar_bc)
print("pot_bc:", pot_bc)
ar_abc = np.arange(8, dtype=np.float64).reshape((2, 2, 2))
pot_abc = Potential(False, [a_node, b_node, c_node], ar_abc)
print("pot_abc:", pot_abc)
ar_bcd = np.arange(0, 80, 10, dtype=np.float64).reshape((2, 2, 2))
pot_bcd = Potential(False, [b_node, c_node, d_node], ar_bcd)
print("pot_bcd:", pot_bcd)
# no need to specify dtype here because using decimal points
ar_ecg = np.array([[[0., 0.60000002], [0., 0.30000001]],
[[0., 0.40000001], [0., 0.69999999]]])
pot_ecg = Potential(False, [e_node, c_node, g_node], ar_ecg)
print("pot_ecg:", pot_ecg)
print("\n-----------------try transpose, distance and ==")
new_abc = cp.deepcopy(pot_abc)
new_abc.set_to_transpose([b_node, a_node, c_node])
print("pot_abc:", pot_abc)
print("new_abc:", new_abc)
assert pot_abc == new_abc
assert pot_abc != (new_abc + 5)
print("distance(pot_abc, new_abc)=",
Potential.distance(new_abc, pot_abc))
print("pot_abc == new_abc?", pot_abc == new_abc)
print("distance(pot_abc, pot_bc)=",
Potential.distance(pot_abc, pot_bc))
print("pot_abc == pot_bc?", pot_abc == pot_bc)
print("\n-----------------try add, sub, mult")
print('pot_ab:', pot_ab)
print('pot_ab2:', pot_ab2)
print("pot_ab + 5:", pot_ab + 5)
print("pot_ab - 5:", pot_ab - 5)
print("pot_ab * 5:", pot_ab * 5)
print("pot_ab + pot_ab2:", pot_ab + pot_ab2)
print("pot_ab - pot_ab2:", pot_ab - pot_ab2)
print("pot_ab * pot_ab2:", pot_ab * pot_ab2)
print("pot_ab + pot_bc:", pot_ab + pot_bc)
print("pot_ab - pot_bc:", pot_ab - pot_bc)
print("pot_ab * pot_bc:", pot_ab * pot_bc)
print("\n-----------------try iadd, isub, and imul")
new_abc = cp.deepcopy(pot_abc)
new_abc.set_to_transpose([b_node, a_node, c_node])
new_abc += 5
assert new_abc == pot_abc + 5
new_abc = cp.deepcopy(pot_abc)
new_abc.set_to_transpose([b_node, a_node, c_node])
new_abc += pot_bc
assert new_abc == pot_abc + pot_bc
new_abc = cp.deepcopy(pot_abc)
new_abc.set_to_transpose([b_node, a_node, c_node])
new_abc -= pot_bc
assert new_abc == pot_abc - pot_bc
new_abc = cp.deepcopy(pot_abc)
new_abc.set_to_transpose([b_node, a_node, c_node])
new_abc *= pot_bc
assert new_abc == pot_abc * pot_bc
print("\n-----------------try truediv")
print("pot_ab/pot_ab2 =", pot_ab / pot_ab2)
print("\n-----------------try itruediv")
new_ab = cp.deepcopy(pot_ab)
new_ab.set_to_transpose([b_node, a_node])
new_ab /= pot_ab2
print("pot_ab/pot_ab2=", new_ab)
assert new_ab == pot_ab / pot_ab2
print("\n-----------------try marginalizing")
new_abc = cp.deepcopy(pot_abc)
print("new_abc=", new_abc)
new_ac = new_abc.get_new_marginal([a_node, c_node])
print("sum_b new_abc=", new_ac)
print("\nnew_abc=", new_abc)
new_ab = new_abc.get_new_marginal([a_node, b_node])
print("sum_c new_abc=", new_ab)
print("\npot_ecg:", pot_ecg)
pot_ge = pot_ecg.get_new_marginal([g_node, e_node])
print("sum_c pot_ecg=", pot_ge)
By an edge we mean a set of two nodes.
Parameters
----------
nd : BayesNode
Returns
-------
list[set[BayesNode]]
"""
edges = list()
neig_list = list(nd.neighbors)
for k in range(len(neig_list)):
n1 = neig_list[k]
for n2 in neig_list[k+1:]:
if not n2.has_neighbor(n1):
edges.append({n1, n2})
return edges
from nodes.Node import *
if __name__ == "__main__":
a = BayesNode(0, name="a")
b = BayesNode(1, name="b")
c = BayesNode(2, name="c")
a.add_neighbor(b)
a.add_neighbor(c)
assert(Star.get_missing_edges_of_node(a) == [{b, c}])
def build_bnet():
"""
Builds QBnet called QuWetGrass with diamond shape
Cloudy
/ \
Rain Sprinkler
\ /
WetGrass
All arrows pointing down
"""
cl = BayesNode(0, name="Cloudy")
sp = BayesNode(1, name="Sprinkler")
ra = BayesNode(2, name="Rain")
we = BayesNode(3, name="WetGrass")
we.add_parent(sp)
we.add_parent(ra)
sp.add_parent(cl)
ra.add_parent(cl)
nodes = {cl, ra, sp, we}
cl.potential = DiscreteUniPot(True, cl) # P(a)
sp.potential = DiscreteCondPot(True, [cl, sp]) # P(b| a)
ra.potential = DiscreteCondPot(True, [cl, ra])
we.potential = DiscreteCondPot(True, [sp, ra, we])
# in general
# DiscreteCondPot(True, [y1, y2, y3, x]) refers to A(x| y1, y2, y3)
# off = 0
# on = 1
cl.potential.pot_arr[:] = [0.5 + 0.1j, 0.5]
ra.potential.pot_arr[1, :] = [0.5 - 0.1j, 0.5 + 0.3j]
ra.potential.pot_arr[0, :] = [0.4, 0.6 - 0.7j]
sp.potential.pot_arr[1, :] = [0.7 + 3.0j, 0.3 - 1.0j]
sp.potential.pot_arr[0, :] = [0.2 + 0.5j, 0.8]
we.potential.pot_arr[1, 1, :] = [0.01 + 1j, 0.99]
we.potential.pot_arr[1, 0, :] = [0.01 - 5.0j, 0.99]
we.potential.pot_arr[0, 1, :] = [0.01, 0.99 + 2.3j]
we.potential.pot_arr[0, 0, :] = [0.99, 0.01 - 0.01j]
cl.potential.normalize_self()
ra.potential.normalize_self()
sp.potential.normalize_self()
we.potential.normalize_self()
return BayesNet(nodes)
elif n == 1 and m == 1:
x = c
y = -theta_hat[3] * s
elif n == 0 and m == 1:
x = theta_hat[2] * s
y = theta_hat[1] * s
elif n == 1 and m == 0:
x = -theta_hat[2] * s
y = theta_hat[1] * s
else:
x = 0
y = 0
s = math.sin(rads[0])
c = math.cos(rads[0])
return (c * x - s * y) + 1j * (c * y + s * x)
if __name__ == "__main__":
pa_nd = BayesNode(0, "pa_nd", size=2)
pa_nd.state_names = ['0', '1']
thetas_degs = [20, 30, 40, 60]
qr = QubitRot(1, "rot", pa_nd, thetas_degs)
print("pa_nd state names: ", pa_nd.state_names)
print("qubit rot state names: ", qr.state_names)
print(qr.potential)
print(qr.potential.get_total_probs())
def build_bnet():
"""
Builds CBnet called WetGrass with diamond shape
Cloudy
/ \
Rain Sprinkler
\ /
WetGrass
All arrows pointing down
"""
cl = BayesNode(0, name="Cloudy")
sp = BayesNode(1, name="Sprinkler")
ra = BayesNode(2, name="Rain")
we = BayesNode(3, name="WetGrass")
we.add_parent(sp)
we.add_parent(ra)
sp.add_parent(cl)
ra.add_parent(cl)
nodes = {cl, ra, sp, we}
cl.potential = DiscreteUniPot(False, cl) # P(a)
sp.potential = DiscreteCondPot(False, [cl, sp]) # P(b| a)
ra.potential = DiscreteCondPot(False, [cl, ra])
we.potential = DiscreteCondPot(False, [sp, ra, we])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
cl.potential.pot_arr[:] = [.5, .5]
ra.potential.pot_arr[1, :] = [.5, .5]
ra.potential.pot_arr[0, :] = [.4, .6]
sp.potential.pot_arr[1, :] = [.7, .3]
sp.potential.pot_arr[0, :] = [.2, .8]
we.potential.pot_arr[1, 1, :] = [.01, .99]
we.potential.pot_arr[1, 0, :] = [.01, .99]
we.potential.pot_arr[0, 1, :] = [.01, .99]
we.potential.pot_arr[0, 0, :] = [.99, .01]
return BayesNet(nodes)
mx : list[int]
my : list[int]
Returns
-------
int
"""
return self.fill_trans_mat_and_st_names_of_nd(
mx, my, dry_run=True)
if __name__ == "__main__":
theta_degs = 35
max_n_sum = 3
in_nd = BayesNode(0, "parent1", size=6)
in_nd.set_state_names_to_product(
[range(2), range(3)], trim=False)
pr = PolarizationRot(1, "pol_rot", in_nd, theta_degs, max_n_sum)
print("in_nd state names: ", in_nd.state_names)
print("pol rot state names: ", pr.state_names)
print(pr.potential)
print("full dict of total probs: ",
pr.potential.get_total_probs())
print("brief dict of total probs: ",
pr.potential.get_total_probs(brief=True))
elif n == 1 and m == 1:
x = c
y = -theta_hat[3]*s
elif n == 0 and m == 1:
x = theta_hat[2]*s
y = theta_hat[1]*s
elif n == 1 and m == 0:
x = -theta_hat[2]*s
y = theta_hat[1]*s
else:
x = 0
y = 0
s = math.sin(rads[0])
c = math.cos(rads[0])
return (c*x - s*y) + 1j*(c*y + s*x)
if __name__ == "__main__":
pa_nd = BayesNode(0, "pa_nd", size=2)
pa_nd.state_names = ['0', '1']
thetas_degs = [20, 30, 40, 60]
qr = QubitRot(1, "rot", pa_nd, thetas_degs)
print("pa_nd state names: ", pa_nd.state_names)
print("qubit rot state names: ", qr.state_names)
print(qr.potential)
print(qr.potential.get_total_probs())
def build_bnet():
"""
Builds QBnet called QuWetGrass with diamond shape
Cloudy
/ \
Rain Sprinkler
\ /
WetGrass
All arrows pointing down
"""
cl = BayesNode(0, name="Cloudy")
sp = BayesNode(1, name="Sprinkler")
ra = BayesNode(2, name="Rain")
we = BayesNode(3, name="WetGrass")
we.add_parent(sp)
we.add_parent(ra)
sp.add_parent(cl)
ra.add_parent(cl)
nodes = {cl, ra, sp, we}
cl.potential = DiscreteUniPot(True, cl) # P(a)
sp.potential = DiscreteCondPot(True, [cl, sp]) # P(b| a)
ra.potential = DiscreteCondPot(True, [cl, ra])
we.potential = DiscreteCondPot(True, [sp, ra, we])
# in general
# DiscreteCondPot(True, [y1, y2, y3, x]) refers to A(x| y1, y2, y3)
# off = 0
# on = 1
cl.potential.pot_arr[:] = [.5 + .1j, .5]
ra.potential.pot_arr[1, :] = [.5 - .1j, .5 + .3j]
ra.potential.pot_arr[0, :] = [.4, .6 - .7j]
sp.potential.pot_arr[1, :] = [.7 + 3.j, .3 - 1.j]
sp.potential.pot_arr[0, :] = [.2 + .5j, .8]
we.potential.pot_arr[1, 1, :] = [.01 + 1j, .99]
we.potential.pot_arr[1, 0, :] = [.01 - 5.j, .99]
we.potential.pot_arr[0, 1, :] = [.01, .99 + 2.3j]
we.potential.pot_arr[0, 0, :] = [.99, .01 - .01j]
cl.potential.normalize_self()
ra.potential.normalize_self()
sp.potential.normalize_self()
we.potential.normalize_self()
return BayesNet(nodes)
def build_bnet():
"""
Builds CBnet in accompanying gif : bnet_HuangDarwiche.gif
From "Inference Belief Networks: A Procedural Guide", by C.Huang and
A. Darwiche
"""
a_node = BayesNode(0, name="A")
b_node = BayesNode(1, name="B")
c_node = BayesNode(2, name="C")
d_node = BayesNode(3, name="D")
e_node = BayesNode(4, name="E")
f_node = BayesNode(5, name="F")
g_node = BayesNode(6, name="G")
h_node = BayesNode(7, name="H")
b_node.add_parent(a_node)
c_node.add_parent(a_node)
d_node.add_parent(b_node)
e_node.add_parent(c_node)
f_node.add_parent(d_node)
f_node.add_parent(e_node)
g_node.add_parent(c_node)
h_node.add_parent(e_node)
h_node.add_parent(g_node)
nodes = {a_node, b_node, c_node, d_node, e_node,
f_node, g_node, h_node}
a_node.potential = DiscreteUniPot(False, a_node) # P(a)
b_node.potential = DiscreteCondPot(False, [a_node, b_node]) # P(b| a)
c_node.potential = DiscreteCondPot(False, [a_node, c_node])
d_node.potential = DiscreteCondPot(False, [b_node, d_node])
e_node.potential = DiscreteCondPot(False, [c_node, e_node])
# P(f|d, e)
f_node.potential = DiscreteCondPot(False, [d_node, e_node, f_node])
g_node.potential = DiscreteCondPot(False, [c_node, g_node])
h_node.potential = DiscreteCondPot(False, [e_node, g_node, h_node])
# in general
# DiscreteCondPot(False, [y1, y2, y3, x]) refers to P(x| y1, y2, y3)
# off = 0
# on = 1
a_node.potential.pot_arr[:] = [.5, .5]
b_node.potential.pot_arr[1, :] = [.5, .5]
b_node.potential.pot_arr[0, :] = [.4, .6]
c_node.potential.pot_arr[1, :] = [.7, .3]
c_node.potential.pot_arr[0, :] = [.2, .8]
d_node.potential.pot_arr[1, :] = [.9, .1]
d_node.potential.pot_arr[0, :] = [.5, .5]
e_node.potential.pot_arr[1, :] = [.3, .7]
e_node.potential.pot_arr[0, :] = [.6, .4]
f_node.potential.pot_arr[1, 1, :] = [.01, .99]
f_node.potential.pot_arr[1, 0, :] = [.01, .99]
f_node.potential.pot_arr[0, 1, :] = [.01, .99]
f_node.potential.pot_arr[0, 0, :] = [.99, .01]
g_node.potential.pot_arr[1, :] = [.8, .2]
g_node.potential.pot_arr[0, :] = [.1, .9]
h_node.potential.pot_arr[1, 1, :] = [.05, .95]
h_node.potential.pot_arr[1, 0, :] = [.95, .05]
h_node.potential.pot_arr[0, 1, :] = [.95, .05]
h_node.potential.pot_arr[0, 0, :] = [.95, .05]
return BayesNet(nodes)
