def setUp(self):
# model a -> b
bn = {}
self.arr = []
bn[0] = BN(domain=Domain(),
new_domain_variables={
'a': [0, 1],
'b': [0, 1]
})
bn[0].add_cpts([
CPT(Factor(variables=['a'], data=[0.5, 0.5]), child='a'),
CPT(Factor(variables=['a', 'b'], data=[0.3, 0.7, 0.4, 0.6]),
child='b')
])
self.arr.append([('a', 'b')])
bn[1] = BN(domain=Domain(),
new_domain_variables={
'a': [0, 1],
'b': [0, 1],
'c': [0, 1]
})
bn[1].add_cpts([
CPT(Factor(variables=['a'], data=[0.5, 0.5]), child='a'),
CPT(Factor(variables=['a', 'b'], data=[0.3, 0.7, 0.4, 0.6]),
child='b'),
CPT(Factor(variables=['c', 'b'], data=[0.1, 0.9, 0.2, 0.8]),
child='c')
])
self.arr.append([('a', 'b'), ('b', 'c')])
self.cbn = [CBN.from_bn(bn[i]) for i in bn.keys()]
def runTest(self):
# results from R version 2.6.2 (2008-02-08) ISBN 3-900051-07-0
r_dx2 = 3453.429
r_dp = 2.2e-16
r_ddf = 4
# regression (NOT known correct! NOT from R)
r_ix2 = 1.08094376504
r_ip = 0.58247332853
r_idf = 2
vars = ['a','b']
vals = {'a': [0,1,2], 'b':[0,1,2]}
ddata = (vars,vals,vars,[(0,0,640),(0,1,1947),(0,2,648),(1,0,1709),(1,1,1364),(1,2,335),(2,0,0),(2,1,3357),(2,2,0)])
iddata = (vars,vals,vars,[(0,0,994),(0,1,2359),(0,2,0),(1,0,1027),(1,1,2312),(1,2,0),(2,0,989),(2,1,2319),(2,2,0)])
x2d = X2Separator(CompactFactor(ddata, domain=Domain()))
x2i = X2Separator(CompactFactor(iddata, domain=Domain()))
dp, dx2, dd = x2d.test_independ('a','b',set())
self.assertEquals(dd,r_ddf)
self.assertAlmostEquals(dx2,r_dx2,3)
self.assertAlmostEquals(dp,r_dp,4)
ip, ix2, id = x2i.test_independ('a','b',set())
self.assertEquals(id,r_idf)
self.assertAlmostEquals(ix2,r_ix2,4)
self.assertAlmostEquals(ip,r_ip,4)
# since there are more zero counts there should be fewer dof.
self.assert_(id < dd)
print ' x2 depend: p =',dp,'x2 =',dx2,'d =',dd
print '!! x2 independ: p =',ip,'x2 =',ix2,'d =',id
# regression (NOT known correct! NOT from R)
r_dg2 = 4224.3017675
r_dp = 0.0
r_ddf = 4
r_ig2 = 1.07917600592
r_ip = 0.582988392395
r_idf = 2
g2d = G2Separator(CompactFactor(ddata, domain=Domain()))
g2i = G2Separator(CompactFactor(iddata, domain=Domain()))
dp, dg2, dd = g2d.test_independ('a','b',set())
self.assertEquals(dd,r_ddf)
self.assertAlmostEquals(dg2,r_dg2,4)
self.assertAlmostEquals(dp,r_dp,4)
ip, ig2, id = g2i.test_independ('a','b',set())
self.assertEquals(id,r_idf)
self.assertAlmostEquals(ig2,r_ig2,4)
self.assertAlmostEquals(ip,r_ip,4)
print '!! g2 depend: p =',dp,'g2 =',dg2,'d =',dd
print '!! g2 independ: p =',ip,'g2 =',ig2,'d =',id
def runTest(self):
# results from R version 2.6.2 (2008-02-08) ISBN 3-900051-07-0
r_dx2 = 109.1547
r_dp = 0
r_ddf = 1
r_ix2 = 0.1236
r_ip = 0.7252
r_idf = 1
vars = ['a','b']
vals = {'a': [0,1], 'b':[0,1]}
ddata = (vars,vals,vars,[(0,0,1509), (0,1,3538), (1,0,1974), (1,1,2979)])
iddata = (vars,vals,vars,[(0,0,1474), (0,1,3610), (1,0,1441), (1,1,3475)])
x2d = X2Separator(CompactFactor(ddata, domain=Domain()))
x2i = X2Separator(CompactFactor(iddata, domain=Domain()))
dp, dx2, dd = x2d.test_independ('a','b',set())
self.assertEquals(dd,r_ddf)
self.assertAlmostEquals(dx2,r_dx2,4)
self.assertAlmostEquals(dp,r_dp,4)
ip, ix2, id = x2i.test_independ('a','b',set())
self.assertEquals(id,r_idf)
self.assertAlmostEquals(ix2,r_ix2,4)
self.assertAlmostEquals(ip,r_ip,4)
print ' x2 depend: p =',dp,'x2 =',dx2,'d =',dd
print ' x2 independ: p =',ip,'x2 =',ix2,'d =',id
# regression (NOT known correct! NOT from R)
r_dg2 = 109.389800878
r_dp = 0
r_ddf = 1
r_ig2 = 0.123551704118
r_ip = 0.725213854926
r_idf = 1
g2d = G2Separator(CompactFactor(ddata, domain=Domain()))
g2i = G2Separator(CompactFactor(iddata, domain=Domain()))
dp, dg2, dd = g2d.test_independ('a','b',set())
self.assertEquals(dd,r_ddf)
self.assertAlmostEquals(dg2,r_dg2,4)
self.assertAlmostEquals(dp,r_dp,4)
ip, ig2, id = g2i.test_independ('a','b',set())
self.assertEquals(id,r_idf)
self.assertAlmostEquals(ig2,r_ig2,4)
self.assertAlmostEquals(ip,r_ip,4)
print '!! g2 depend: p =',dp,'g2 =',dg2,'d =',dd
print '!! g2 independ: p =',ip,'g2 =',ig2,'d =',id
def setUp(self):
from gPy.Variables import Domain
self.bnm = BN(domain=Domain()) # don't use default domain
self.bnm.from_dnet(read_dnet('Asia.dnet'))
self.cptdict = self.bnm.factors
# taken directly from Netica output
self.marginals = [
Factor((('VisitAsia'), ), [0.99, 0.01]),
Factor((('Tuberculosis'), ), [0.9896, 0.0104]),
Factor((('Smoking'), ), [0.5, 0.5]),
Factor((('Cancer'), ), [0.945, 0.055]),
Factor((('TbOrCa'), ), [0.93517, 0.064828]),
Factor((('XRay'), ), [0.11029, 0.88971]),
Factor((('Bronchitis'), ), [0.55, 0.45]),
Factor((('Dyspnea'), ), [0.56403, 0.43597])
]
# taken directly from Netica output
self.cond_marginals = [
Factor((('VisitAsia'), ), [0.95192, 0.048077]),
Factor((('Tuberculosis'), ), [0, 1]),
Factor((('Smoking'), ), [0.52381, 0.47619]),
#other marginals are conditional on these values
#Factor((('Cancer'),),
# [1,0]),
#Factor((('TbOrCa'),),
# [0,1]),
Factor((('XRay'), ), [0.98, 0.02]),
Factor((('Bronchitis'), ), [0.55714, 0.44286]),
Factor((('Dyspnea'), ), [0.21143, 0.78857])
]
def runTest(self):
samples = GibbsSampler(self._minibn_do).samples(100000)
data = CompactFactor(samples,domain=Domain())
p = self._minibn_do.copy(copy_domain=True)
p.estimate_parameters(data)
self.failUnless(same_factor(distribution_of(p), distribution_of(self._minibn_do), dp=2, verbose=True))
def runTest(self):
# construct factors of various sizes with no data
for sz in xrange(6):
vars = ['V'+str(i) for i in xrange(sz)]
vals = dict([(v,[0,1]) for v in vars])
data = (vars,vals,vars,[])
for v_on in subsetn(vars, sz):
inst = []
for v in vars:
if v in v_on:
inst.append(1)
else:
inst.append(0)
data[3].append(tuple(inst+[0]))
d = CompactFactor(data,domain=Domain())
x2 = X2Separator(d)
g2 = G2Separator(d)
for a,b in pairs(vars):
for s in powerset(set(vars) - set([a,b])):
x2p, x2s, x2d = x2.test_independ(a, b, set(s))
g2p, g2s, g2d = g2.test_independ(a, b, set(s))
# one degree of freedom
self.assertEquals(x2d, 0)
self.assertEquals(g2d, 0)
# default to independent
self.assertEquals(x2p, 1)
self.assertEquals(g2p, 1)
# zero statistics (no data)
self.assertEquals(x2s, 0)
self.assertEquals(g2s, 0)
def rand_factor(vs):
n = reduce(operator.mul, [len(vs[v]) for v in vs])
f = Factor(variables=vs.keys(),
data=rand_factor_data(n),
domain=Domain(),
check=True,
new_domain_variables=vs)
return f
def runTest(self):
data = CompactFactor(read_csv(open('tetrad_asia.csv')),domain=Domain())
ci = PCCI(G2Separator(data))
g = ICPattern(ci)
self.assertEquals(g.shd(self._asia_pdag),5)
self.assertEquals(self._tetrad_pdag.shd(self._asia_pdag),4)
# I think tetrad is wrong (in terms of implementation)
self.assertEquals(g.shd(self._tetrad_pdag),1)
def runTest(self):
num_vars = 5
num_vals = 10
for x in xrange(num_runs):
vs = dict([('V'+str(i),range(num_vals)) for i in xrange(num_vars)])
# should get some random data with 0.
f = Factor(variables = vs.keys()
,data = [abs(randint(0,20)) for i in xrange(num_vals**num_vars)]
#,data = [5, 0, 1, 2]
#,data = [13,0,0,20]
#,data = [14,15,20,16]
,domain = Domain()
,new_domain_variables=vs
,check = True)
records = []
for inst in f.insts():
records.append(inst + (f[inst],))
rawdata = (vs.keys(), vs, vs.keys(), records)
#print records
cf = IncrementalCompactFactor(rawdata, rmin=0)
#print 'old tree:'
#print cf
# g = Factor(variables = vs.keys()
# #,data = [13,2,0,0]
# ,data = [14,15,0,0]
# ,domain = Domain()
# ,new_domain_variables=vs
# ,check = True)
g = f.copy(copy_domain=True)
def swap_some(x):
r = random()
if x == 0:
if r <= 0.5:
return randint(1,5)
return 0
elif r <= 0.5:
return 0
return x
def invert_some(x):
return 5-x
g.map(swap_some)
#print g
f += g
records = []
for inst in g.insts():
records.append(inst + (g[inst],))
rawdata = (vs.keys(), vs, vs.keys(), records)
cf.update(rawdata)
#print 'new tree:'
#print cf
for variables in powerset(vs.keys()):
g = f.copy(copy_domain=True)
g.marginalise_away(g.variables() - frozenset(variables))
self.assert_(same_factor(g,cf.makeFactor(variables),verbose=True))
def testdnet(self):
from gPy.IO import read_dnet
from gPy.Models import BN
from gPy.Variables import Domain
bnm = BN(domain=Domain())
bnm.from_dnet(read_dnet('Asia.dnet'))
self.samegraph(bnm.adg(), self.asia_adg)
for name, cpt_in_file in self.asia_cpts.items():
cpt = bnm[name]
self.samecpt(cpt, cpt_in_file, cpt.child())
def runTest(self):
data = CompactFactor(read_csv(open('tetrad_xor.csv')),domain=Domain())
ci = PCCI(G2Separator(data))
print ci._ind
for a,b in pairs(data.variables()):
if a == 'X1' and b == 'X2' or a == 'X2' and b == 'X1':
self.assert_(ci.has_independence(a, b))
self.assert_(not ci.has_independence_involving(a,b,'X3'))
else:
print a,b
self.assert_(not ci.has_independence(a,b))
data = CompactFactor(read_csv(open('tetrad_xor.csv')),domain=Domain())
ci = PCCI(G2Separator(data))
for a,b in pairs(data.variables()):
if a == 'X1' and b == 'X2' or a == 'X2' and b == 'X1':
self.assert_(ci.has_independence(a, b))
self.assert_(not ci.has_independence_involving(a,b,'X3'))
else:
print a,b
self.assert_(not ci.has_independence(a,b))
def tryModel(self, model):
# generate some samples
cf = []
sampler = ForwardSampler(model)
samples = sampler.samples(100)
cf.append(CompactFactor(samples,domain=Domain()))
icf = IncrementalCompactFactor(samples, domain=Domain())
for i in xrange(10):
samples = sampler.samples(100)
cf.append(CompactFactor(samples,domain=Domain()))
icf.update(samples)
# see if the sum of the CPT Factors in cf match that of icf
for child in model.variables():
family = model.adg().parents(child) | set([child])
a = icf.makeFactor(family)
b = cf[0].makeFactor(family)
for f in cf[1:]:
b += f.makeFactor(family)
self.assert_(same_factor(a,b))
def observe(self, num_samples, skip=0):
"""Draw C{num_samples} observational samples, separated by C{skip}
steps of the sampler. These samples are appended to any existing
observational samples and then returned.
@return: L{IncrementalCompactFactor}
"""
samples = self._pure_sampler.samples(num_samples,skip)
if self._data is None:
self._data = IncrementalCompactFactor(samples,domain=Domain())
else:
self._data.update(samples)
return self._data
def __init__(self, bn,burnin=1000):
"""
@param bn: A causal Bayesian network from which samples are drawn.
@type bn: L{CBN}
@param burnin: The burn in for the L{GibbsSampler} used for generating
interventional data.
@type burnin: int
"""
super(CausalWorld,self).__init__(domain=Domain.copy(bn),variables=bn.variables())
self._pure_model = bn.copy(copy_domain=True)
self._pure_sampler = ForwardSampler(self._pure_model)
self._burnin = burnin
self._data = None
self._inter_sampler = {}
self._inter_data = {}
def rand_bn(vs, max_potential_parents=15):
model = BN(domain=Domain(), new_domain_variables=vs)
for child in vs.keys():
parents = list(model.variables())
too_many = len(parents) - max_potential_parents
if too_many > 0:
for i in xrange(too_many):
parents.remove(choice(parents))
fv = rand_subset(parents) | set([child])
n = reduce(operator.mul, [len(vs[v]) for v in fv])
f = Factor(variables=fv,
data=rand_factor_data(n),
domain=model,
check=True)
cpt = CPT(f, child, True, True)
model *= cpt
return model
def runTest(self):
num_vars = 5
num_vals = 10
for i in xrange(num_runs):
vs = dict([('V'+str(i),range(num_vals)) for i in xrange(num_vars)])
# should get some random data with 0.
f = Factor(variables = vs.keys()
,data = [abs(randint(0,5)) for i in xrange(num_vals**num_vars)]
,domain = Domain()
,new_domain_variables=vs
,check = True)
records = []
for inst in f.insts():
records.append(inst + (f[inst],))
rawdata = (vs.keys(), vs, vs.keys(), records)
cf = IncrementalCompactFactor(rawdata)
for variables in powerset(vs.keys()):
g = f.copy(copy_domain=True)
g.marginalise_away(g.variables() - frozenset(variables))
self.assert_(same_factor(g,cf.makeFactor(variables),verbose=True))
def generate_dense_bn(density, num_vars=8, num_vals=3):
if density > num_vars:
raise RuntimeError, 'density must be less than number of variables'
vars, parents = generate_dense_parents(density, num_vars)
vals = dict([(var, frozenset([i for i in xrange(num_vals)]))
for var in vars])
bn = BN(domain=Domain(), new_domain_variables=vals)
for child in vars:
if child in parents:
n = num_vals**(len(parents[child]) + 1)
else:
n = num_vals
parents[child] = frozenset()
f = Factor(variables=frozenset([child]) | parents[child],
data=rand_factor_data(n),
domain=bn,
check=True)
bn *= CPT(f, child, True, True)
return bn
def query(self, intervention, num_samples,skip=0):
"""Draw C{num_samples} interventional samples, separated by C{skip}
steps of the sampler. These samples are appended to any existing
interventional samples and then returned. The intervention made is
that of C{intervention}.
@param intervention: A dictionary mapping variables in the L{CBN} to
a single value in the domain.
@return: L{IncrementalCompactFactor}
"""
k = frozenset(intervention.keys())
if not self._inter_sampler.has_key(k):
do_model = CBN.from_bn(self._pure_model.copy(copy_domain=True))
do_model.intervene(intervention)
self._inter_sampler[k] = GibbsSampler(do_model, self._burnin)
do_sampler = self._inter_sampler[k]
samples = do_sampler.samples(num_samples,skip)
if not self._inter_data.has_key(k):
self._inter_data[k] = IncrementalCompactFactor(samples,domain=Domain())
else:
self._inter_data[k].update(samples)
return self._inter_data[k]
def rand_fr(vs, min_fact=1, max_fact=10, min_fact_vars=1, max_fact_vars=10):
model = FR(domain=Domain(), new_domain_variables=vs)
for i in xrange(randrange(min_fact, max_fact)):
fv = []
while len(fv) == 0:
for j in xrange(
randrange(min_fact_vars, min(max_fact_vars,
len(vs.keys())))):
v = choice(vs.keys())
while v in fv:
v = choice(vs.keys())
fv.append(v)
fv = tuple(fv)
n = reduce(operator.mul, [len(vs[v]) for v in fv])
f = Factor(variables=fv,
data=rand_factor_data(n),
domain=model,
check=True)
model *= f
return model
def setUp(self):
from gPy.Variables import Domain
bnm = BN(domain=Domain())
bnm.from_dnet(read_dnet('Asia.dnet'))
self.hypergraph = bnm._hypergraph
self.adg = bnm._adg
self.tarjan = UGraph(range(1,11),
((1,2),(1,3),(2,3),(2,10),(3,10),(4,5),
(4,7),(5,6),(5,9),(5,7),(6,7),(6,9),
(7,8),(7,9),(8,9),(8,10),(9,10)))
self.tarjan2 = UGraph(range(1,10),
((1,4),(1,3),(2,3),(2,7),(3,5),(3,6),
(4,5),(4,8),(5,6),(5,8),(6,7),(6,9),
(7,9),(8,9)))
self.tarjan3 = UGraph(range(1,10),
((1,4),(1,3),(2,3),(2,7),(3,5),(3,6),
(4,5),(4,8),(5,6),(5,8),(6,7),(6,9),
(7,9),(8,9),
(3,4),(3,7),(4,6),(4,7),(5,7),(6,8),(7,8)))
self.tarjanh1 = Hypergraph([[3,4],[2,4],[1,2,3]])
self.tarjanh2 = Hypergraph([[3,4],[2,4],[1,2,3],[2,3,4]])
self.graph1 = UGraph('ABCDEF',('AB','AC','BD','CE','EF'))
self.graph2 = UGraph('ABCDEF',('AB','AC','BD','CE','EF','BC','CD','DE'))
from gPy.Examples import minibn, asia
from gPy.Models import FR, BN
from gPy.Parameters import Factor, CPT
from gPy.Variables import Domain
from random import choice, randrange, uniform, shuffle
import operator, unittest, pickle
xor = BN(domain=Domain(),
new_domain_variables={
'a': [0, 1],
'b': [0, 1],
'c': [0, 1]
})
xor.add_cpts([
CPT(Factor(variables=['a'], data=[0.5, 0.5]), child='a'),
CPT(Factor(variables=['b'], data=[0.5, 0.5]), child='b'),
CPT(Factor(variables=['c', 'a', 'b'],
data=[1.0, 0.0, 0.0, 1.0, 0.0, 1.0, 1.0, 0.0]),
child='c')
])
cbn_small_names = ['xor', 'minibn', 'asia']
cbn_small_test_cases = [xor, minibn, asia]
cbn_large_names = ['alarm', 'insurance', 'carpo']
try:
# load the pickled large Bayes nets.
cbn_large_test_cases = map(
lambda fn: pickle.load(open('networks/' + fn + '_bn.pck', 'r')),
cbn_large_names)
except:
cbn_large_names = []
cbn_large_test_cases = []
def disp(fn, samples):
f = open(fn, 'w')
fact = samples.makeFactor(samples.variables())
for var in fact.variables():
print >> f, var,
print >> f, 'count'
for inst in fact.insts():
for i in inst:
print >> f, i,
print >> f, fact[inst]
f.close()
bn0 = BN(domain=Domain(), new_domain_variables={'a': [0, 1], 'b': [0, 1]})
bn0.add_cpts([
CPT(Factor(variables=['a'], data=[0.5, 0.5]), child='a'),
CPT(Factor(variables=['a', 'b'], data=[0.3, 0.7, 0.4, 0.6]), child='b')
])
w = CausalWorld(bn0)
samples = w.observe(10000)
disp('two_depend', samples)
bn1 = BN(domain=Domain(), new_domain_variables={'a': [0, 1], 'b': [0, 1]})
bn1.add_cpts([
CPT(Factor(variables=['a'], data=[0.5, 0.5]), child='a'),
CPT(Factor(variables=['b'], data=[0.3, 0.7]), child='b')
])
w = CausalWorld(bn1)
samples = w.observe(10000)