def do_individual_experiment(args):
lookahead, seed = args
heur = npuzzle.ManhattanDistanceHeuristic(False)
goal = npuzzle.NPuzzleBoard.done
rta = search.Rta(heur=heur, goal=goal, lookahead=lookahead, learn=False)
env = npuzzle.NPuzzle(n1sqrt=n1sqrt, difficulty=0.5, seed=seed)
done, elapsed = rta(env, stats=True)
return env.steps(), elapsed
def manual_play():
env = npuzzle.NPuzzle(n1sqrt=n1sqrt, difficulty=0.5)
env.render()
while True:
print("---")
print_usage()
cmd = int(input("Cmd: "))
if cmd == -1:
break
done = env.step(cmd)
env.render()
if done:
print("done in {} step(s)!".format(env.steps()))
break
import npuzzle
p = npuzzle.NPuzzle(3)
p.visualise()
tile = p.read_tile(1, 2)
print(tile)
p.reset()
p.visualise()
try:
p.read_tile(3, 3)
except IndexError:
print('index eror test passed')
return True
if odd_even_inv == 1 and odd_even_space_idx == 0:
return True
else:
return False
except:
for i in range(3):
for j in range(3):
if env.read_tile(i, j) is None:
env_line.append(9)
else:
env_line.append(env.read_tile(i, j))
for i in range(len(env_line)):
for j in range(i + 1, len(env_line)):
if env_line[i] > env_line[j]:
count += 1
if count % 2 == 0:
odd_even_inv = 0 # inversion is even
return True
else:
odd_even_inv = 1 # inversion is odd
return False
if __name__ == "__main__":
env = npuzzle.NPuzzle(4) # env = npuzzle.NPuzzle(4)
env.reset()
env.visualise()
# just check
print(is_solvable(env))
def __blank_row_number_from_bottom(tiles, size):
index = tiles.index(None)
row = index // size
row_decs = size - row
return row_decs
def __is_even(N):
return N % 2 == 0
def __inversion_count(tiles, size):
inversion_count = 0
for i in range(size * size - 1):
n = tiles[i]
for j in range(i, size * size - 1):
n_next = tiles[j + 1]
if n_next is not None and n is not None and n > n_next:
inversion_count += 1
return inversion_count
if __name__ == "__main__": # testing suite
env = npuzzle.NPuzzle(3) # instance of NPuzzle class
env.reset() # random shuffle
env.visualise() # just to show the tiles
# just check
print(is_solvable(env)) # should output True or False<Paste>
def astar(seed):
env = npuzzle.NPuzzle(n1sqrt=n1sqrt, difficulty=0.5, seed=seed)
heur = npuzzle.ManhattanDistanceHeuristic(False)
goal = npuzzle.NPuzzleBoard.done
plan, _, elapsed = search.astar(env.state(), heur, goal, True)
return len(plan), elapsed