[Python] Python multiprocessing, Knight's Tour [closed]

I am trying to brute-force calculate all possible paths for a knight's tour.

num_size = 5
path_list = []

DirX = [2, 1, -1, -2, -2, -1, 1, 2]
DirY = [1, 2, 2, 1, -1, -2, -2, -1]

def isSafe(i, j, n, Board):
return i >= 0 and j >= 0 and i < n and j < n and Board[j] == 0

isPossible = False

def knightTour(ChessBoard, path, N, x, y, visited=1):
global isPossible

ChessBoard[x][y] = visited

if visited==N*N :
isPossible = True
global path_list
path_list.append(path)

ChessBoard[x][y] = 0
return

for i in range(8):
newX = x + DirX
newY = y + DirY
if isSafe(newX, newY, N, ChessBoard) and not ChessBoard[newX][newY]:
knightTour(ChessBoard, path+[(newX,newY)], N, newX, newY, visited + 1)

ChessBoard[x][y] = 0

if __name__ == "__main__":
ChessBoard = np.array([[0 for j in range(num_size)] for i in range(num_size)])
knightTour(ChessBoard, [(0,0)], num_size, 0, 0)


I thought I could use multiprocessing to speed up the knightTour function but after having refactored the moves into a function on the outer level, multiprocessing.Pool and ProcessPoolExecutor gave me errors and ThreadPoolExecutor didn't even start. What is the solution?

Continue reading...