替代Python中的循环嵌套

Question

I've read that one of the key beliefs of Python is that flat > nested. However, if I have several variables counting up, what is the alternative to multiple for loops? My code is for counting grid sums and goes as follows:

def horizontal():
    for x in range(20):
        for y in range(17):
            temp = grid[x][y: y + 4]
            sum = 0
            for n in temp:
                sum += int(n)
            print sum # EDIT: the return instead of print was a mistype

This seems to me like it is too heavily nested. Firstly, what is considered to many nested loops in Python ( I have certainly seen 2 nested loops before). Secondly, if this is too heavily nested, what is an alternative way to write this code?

Answer 1

from itertools import product

def horizontal():
    for x, y in product(range(20), range(17)):
        print 1 + sum(int(n) for n in grid[x][y: y + 4])

You should be using the sum function. Of course you can't if you shadow it with a variable, so I changed it to my_sum

Answer 2

grid = [range(20) for i in range(20)]
sum(sum( 1 + sum(grid[x][y: y + 4]) for y in range(17)) for x in range(20))

The above outputs 13260, for the particular grid created in the first line of code. It uses sum() three times. The innermost sum adds up the numbers in grid[x][y: y + 4] , plus the slightly strange initial value sum = 1 shown in the code in the question. The middle sum adds up those values for the 17 possible y values. The outer sum adds up the middle values over possible x values.

If elements of grid are strings instead of numbers, replace
sum(grid[x][y: y + 4])
with
sum(int(n) for n in grid[x][y: y + 4]

Answer 3

You can use a dictionary to optimize performance significantly

This is another example:

locations = {}
for i in range(len(airports)):
    locations[airports["abb"][i][1:-1]] = (airports["height"][i], airports["width"][i])

for i in range(len(uniqueData)):
    h, w = locations[uniqueData["dept_apt"][i]]
    uniqueData["dept_apt_height"][i] = h
    uniqueData["dept_apt_width"][i] = w