How can I speed up my misplaced tiles heuristic for the 8 puzzle problem?

Question

My lists are always of length 8 (7 indices), and always contain numbers 0-8

I currently do this to find the sum of misplaced tiles:

def misplacedTilesHeuristic(stateObj, goal):
    sum = 0

    for elem in range(len(goal)):
        if goal[elem] != stateObj[elem]:
            sum+=1

    return sum

How can I make this faster?

Edit:

misplacedTilesHeuristic((4, 5, 3, 1, 0, 6, 7, 2, 8), (0, 1, 2, 3, 4, 5, 6, 7, 8))

Asked By: Aeryes

||

Source

Answer 1

as already mentioned, the one-liner is a good idea, for example like this :

def comp(stObj,goal):
    sum = 0
    for elem in range(len(goal)):
        if goal[elem] != stObj[elem]:sum +=1
    return sum

def prop1(stObj,goal):
    sum = 0
    for i,j in zip(stObj,goal):
        if i !=j:sum +=1
    return sum

def prop2(stObj,goal):
    return sum([i!=j for i, j in zip(stObj,goal)])

def prop3(stObj,goal):
    return sum([i is not j for i, j in zip(stObj,goal)])

def prop4(stObj,goal):
    return sum(map(lambda x, y: x != y, stObj, goal))

t = (4, 5, 3, 1, 0, 6, 7, 2, 8), (0, 1, 2, 3, 4, 5, 6, 7, 8)

Benchmark:

%timeit comp(*t)
1.64 µs ± 46.9 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)
%timeit prop1(*t)
1.22 µs ± 27.9 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each) 
%timeit prop2(*t)
1.67 µs ± 86.5 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)
%timeit prop3(*t)
1.6 µs ± 48.3 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)
%timeit prop4(*t)
1.79 µs ± 32.4 ns per loop (mean ± std. dev. of 7 runs, 1000000 loops each)

The prop1() shows the best times for the moment with almost 34.4% better performance, but I think that it could be better 🙂

Answered By: baskettaz

Answer 2

According to my information your code is best way to compute misplaced tiles h1 but there is a little mistake in your code that needs correction.

h1 is number of misplaced tiles except 0. your code is not handling this condition

def misplaced_tiles(state, goal):
 h1 = 0
 for item in range(len(goal)):
   if stat[item] != 0 :
      if state[item] != goal[item]:
          h1 += 1

 return h1

I hope you will get my point -:)

Answered By: eitzaz shah

How can I speed up my misplaced tiles heuristic for the 8 puzzle problem?

Question:

Answers: