def __init__(self, moves = None):
if moves is None:
moves = [Move.generate_random_move()
for i in xrange(NUMBER_OF_MOVES_PER_MOVE_LIST)]
list.__init__(self, moves)
return getattr(self, attr)
if __name__ == "__main__":
log = open("validation.log", "w")
distance_to_numerical_log = open("dtn.log", "w")
TRIALS = 50
for radiation in [1, 3, 10, 50, 100, 200, 500, 1000]:
print "Radiation", radiation
log.write("Radiation Strength: %d\n" % radiation)
trial_results = np.zeros(TRIALS, dtype=np.float64)
trial_distances = np.zeros(TRIALS, dtype=np.float64)
print "Trial",
for trial in xrange(TRIALS):
log.write("\tTrial: %d\n" % trial)
original = Move.generate_random_move()
mutant = original.copy()
for x in xrange(radiation):
mutant = mutant.mutate()
prior_time = time.time()
results = np.zeros(10 ** 6, dtype=np.float64)
for i, row in enumerate(itertools.product(xrange(0, 100, 10), repeat=6)):
s = state(row)
results[i] = original.evaluate(s) - mutant.evaluate(s)
# print sp.pearsonr(original_values, mutant_values)
d = distance(original, mutant)
deviation = np.std(results, dtype=np.float64)
log.write("\t\tOriginal: %s\n" % (original,))
log.write("\t\tMutant: %s\n" % (mutant,))
log.write("\t\tDistance: %d\n" % (d,))
log.write("\t\tNumerical Deviation %f\n" % (deviation,))
