def __init__(self, cheat=False):
self.cheat = cheat
if cheat is True:
self.database = AvianAsker.initialize()
else:
self.database = None
self.initialize()
import AvianAsker as AA
import numpy as np
# NOTE: good idea to remove 'picklefile' if this is the first time running this script (or using updated AvianAsker code)
database = AA.initialize()
# make an unknown bird
unkbird = np.ones(database.num_attributes)*0.5
# set some random parameters:
data = [[1,1]]
for datum in data:
unkbird[datum[0]] = datum[1]
# fill out related attributes:
unkbird = AA.fillout_related_attributes(database, unkbird)
# remove impossible birds:
AA.remove_impossible_birds(database, unkbird)
# print number of remaining birds:
print 'remaining possible birds: ', AA.count_permissible_birds(database)
# calculate probabilities for unknown attributes:
p = AA.calc_attribute_probabilities(database, unkbird)
print 'best guess at unkbird attributes: '
print p
