def __init__(self, hypos):
Suite.__init__(self)
for hypo in hypos:
if(hypo == 'No_Cancer'):
self.Set(hypo, .99)
else:
self.Set(hypo, .01)
def __init__(self, inferred_goal, hypotheses, actions, \
prob_non_compliance=0.1, prob_viol_detection=0.5, \
prob_sanctioning=0.2, prob_random_punishment=0.01, \
name=''):
Suite.__init__(self, hypotheses, name)
self.initial_hypotheses = hypotheses # Keep the initial hypotheses to allow later reinitilization
self.inferred_goal = inferred_goal
self.actions = actions
self.nodes = {node for action in actions for node in action.path}
self.prob_non_compliance = prob_non_compliance
self.prob_viol_detection = prob_viol_detection
self.prob_sanctioning = prob_sanctioning
self.prob_random_punishment = prob_random_punishment
self.SetGoal(inferred_goal)
def __init__(self, prior_type='uniform'):
Suite.__init__(self)
rangemin, rangemax = 0, 101
if prior_type == 'uniform':
priors = map(lambda x: (x, 1), range(rangemin, rangemax))
elif prior_type == 'triangle':
rangemid = ceil((rangemin + rangemax) / 2)
priors = map(
lambda x: (x, x) if x < rangemid else (x, rangemax - x),
range(rangemin, rangemax))
for val, prob in priors:
self.Set(val, prob)
self.Normalize()
def __init__(self,
inferred_goal,
hypotheses,
actions,
prob_non_compliance=0.1,
name=''):
Suite.__init__(self, hypotheses, name)
self.inferred_goal = inferred_goal
self.actions = actions
self.prob_non_compliance = prob_non_compliance
if planned(inferred_goal, actions):
self.plans = flattened_plan_tree(inferred_goal)
else:
print "Error: Failed to find any plans for %s given actions %s" % (
inferred_goal, actions)
def __init__(self, hypos):
Suite.__init__(self)
def __init__ (self, hypos, alpha = 1.0):
Suite.__init__(self)
for hypo in hypos:
self.Set(hypo, hypo ** (-alpha))
self.Normalize()
def __init__(self):
Suite.__init__(self, xrange(0, 101))
# for x in xrange(1, 101):
# self.Set(x, 1)
# self.Normalize()
self.real_val = thinkbayes.Beta().Random()
