>>> cpt = burglary.variable_node('Alarm').cpt
>>> parents = ['Burglary', 'Earthquake']
>>> event = {'Burglary': True, 'Earthquake': True}
>>> print '%4.2f' % cpt.p(True, parents, event)
0.95
>>> event = {'Burglary': False, 'Earthquake': True}
>>> print '%4.2f' % cpt.p(False, parents, event)
0.71
>>> BoolCPT({T: 0.2, F: 0.625}).p(False, ['Burglary'], event)
0.375
>>> BoolCPT(0.75).p(False, [], {})
0.25

(fixme: The following test p_values which has been folded into p().)
>>> cpt = BoolCPT(0.25)
>>> cpt.p_values(F, ())
0.75
>>> cpt = BoolCPT({T: 0.25, F: 0.625})
>>> cpt.p_values(T, (T,))
0.25
>>> cpt.p_values(F, (F,))
0.375
>>> cpt = BoolCPT({(T, T): 0.2, (T, F): 0.31,
...  (F, T): 0.5, (F, F): 0.62})
>>> cpt.p_values(T, (T, F))
0.31
>>> cpt.p_values(F, (F, F))
0.38


>>> cpt = BoolCPT({True: 0.2, False: 0.7})
>>> cpt.rand(['A'], {'A': True}) in [True, False]
True
>>> cpt = BoolCPT({(True, True): 0.1, (True, False): 0.3,
...   (False, True): 0.5, (False, False): 0.7})
>>> cpt.rand(['A', 'B'], {'A': True, 'B': False}) in [True, False]
True


>>> enumeration_ask('Earthquake', {}, burglary).show_approx()
'False: 0.998, True: 0.002'


>>> s = prior_sample(burglary)
>>> s['Burglary'] in [True, False]
True
>>> s['Alarm'] in [True, False]
True
>>> s['JohnCalls'] in [True, False]
True
>>> len(s)
5


>>> s = {'A': True, 'B': False, 'C': True, 'D': False}
>>> consistent_with(s, {})
True
>>> consistent_with(s, s)
True
>>> consistent_with(s, {'A': False})
False
>>> consistent_with(s, {'D': True})
False

>>> seed(21); p = rejection_sampling('Earthquake', {}, burglary, 1000)
>>> [p[True], p[False]]
[0.001, 0.999]

>>> seed(71); p = likelihood_weighting('Earthquake', {}, burglary, 1000)
>>> [p[True], p[False]]
[0.002, 0.998]