Python: modifying list with nested loops in list comprehension

Question

I have part of a code that does something analogous to the following:

import numpy as np
N = 10
test = [1] * N
np.random.seed(123)
norm = np.random.rand(N)
my_dic = {(k, kk):np.random.rand(1) for k,_ in enumerate(test) for kk in range(5)}
for i, _ in enumerate(test):
    test[i] *= norm[i]
    for j in range(5):
        test[i] *= my_dic[(i, j)]

Since both loop are modifying a list, is there a way for me to translate that to list comprehension? I have been trying variations of the following, with no success:

import numpy as np
N = 10
test = [1] * N
np.random.seed(123)
norm = np.random.rand(N)
my_dic = {(k, kk):np.random.rand(1) for k,_ in enumerate(test) for kk in range(5)}

test = [val for val in norm]
test = [test[i] * my_dic[(i, j)] for i, _ in enumerate(test) for j in range(5)]

List comprehensions are for creating new lists, not for modifying existing lists. if you want to modify a list in-place, the idiomatic way is to just use a for-loop. — juanpa.arrivillaga
– juanpa.arrivillaga, Commented Oct 3, 2018 at 18:18

zwer · Accepted Answer · 2018-10-03 18:29:05Z

1

If you insist doing it with a single list comprehension, I guess you can do something like:

from operator import mul
from functools import reduce  # comment out on Python 2.x

test[:] = [reduce(mul, [v * norm[i]] + [my_dic[(i, j)] for j in range(5)], 1)
           for i, v in enumerate(test)]

Or:

test[:] = [v * norm[i] * reduce(mul, (my_dic[(i, j)] for j in range(5)), 1)
           for i, v in enumerate(test)]

But the real question is - why? You're making it harder to read and/or maintain and you're not getting much in the performance department.

answered Oct 3, 2018 at 18:29

zwer

25.9k3 gold badges53 silver badges70 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Python: modifying list with nested loops in list comprehension

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related