def show_policy(meth):
policy = Policy(laby, meth)
names = {(1, 0): 'R', (-1, 0): 'L', (0, 1): 'D', (0, -1): 'U'}
for y in xrange(laby.h):
for x in xrange(laby.w):
state = LabyState((x, y))
if laby.board[y][x] == '#' or laby.board[y][x] == 'G':
print laby.board[y][x],
else:
actions = policy.actions_probas_from(state)
if len(actions) > 0:
for action, proba in actions:
if proba > 0:
print names[action],
break
else:
print ' ',
print
