初始化字典中键/值对的最佳方法

Question

I'm working in Python 3

I have a board with cells that I have to store, each cell has a coordinate and needs to store 3 values. I used a dictionary where the key equals the coordinates of the cell and the value is a list of 3 elements that stores the required values. Will it be fast to get and read these values when the board gets big and a lot of key/value pairs are stored? Or what other method would be more efficient in terms of speed?

Alex

EDIT:

The 3 values that I need for each cell are zero's or ones. I have a tuple that contains the dimension of the board, a tuple that contains all the cell positions that should have a value of 1 as a first value, a frozenset that contains all the cell positions that should have a value of 1 as a second value and a tuple that contains all the cell positions that should have a value of 1 as a 3rd value:

tuple1 is a tuple that contains tuples that each represent the positions of the cells that have a 1 as first value, frozenset contains tuples that each represent the positions of the cells that have a 1 as second value, and tuple2 contains tuples that each represent the positions of the cells that have 1 as a third value (in the list)

Code:

board = {}

for row in range(dimension[0]):
    for col in range(dimension[1]):
        board[(row,col)] = [0,0,0] #default
for x in tuple1:
    board[x][0] = 1

for x in frozenset:
    board[x][1] = 1

for x in tuple2:
    board[x][2] = 1

return board

Will it now be fast to look up keys and there values?

EDIT2: is this the fastest method to set up a dictionary like this?

Answer 1

With numpy

If you're looking for performance, I think you should seriously consider numpy with boolean values. Your 2-D Board with tuples as values could be seen as a 3D Matrix:

import numpy as np

width, height = 5, 5

board = np.full((width,height,3), False, dtype = bool)
tuple1 = ((1,2), (3,4), (0,2))

for i,j in tuple1:
    board[i,j,0] = True

print(board)
#   [[[False False False]
#     [False False False]
#     [ True False False]
#     [False False False]
#     [False False False]]

#    [[False False False]
#     [False False False]
#     [ True False False]
#     [False False False]
#     [False False False]]

#    [[False False False]
#     [False False False]
#     [False False False]
#     [False False False]
#     [False False False]]

#    [[False False False]
#     [False False False]
#     [False False False]
#     [False False False]
#     [ True False False]]

#    [[False False False]
#     [False False False]
#     [False False False]
#     [False False False]
#     [False False False]]]

With standard lists

If you cannot use numpy , you could simply use standard lists:

width, height = 5, 5

board = [[[0, 0, 0] for j in range(width)] for i in range(height)]
tuple1 = ((1,2), (3,4), (0,2))

for i,j in tuple1:
    board[i][j][0] = 1

from pprint import pprint
pprint(board)  
[[[0, 0, 0], [0, 0, 0], [1, 0, 0], [0, 0, 0], [0, 0, 0]],
 [[0, 0, 0], [0, 0, 0], [1, 0, 0], [0, 0, 0], [0, 0, 0]],
 [[0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0]],
 [[0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0], [1, 0, 0]],
 [[0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0], [0, 0, 0]]]

I don't think there's any advantage in using a dict. It would just shuffle your cells and make display harder.