How can I change an element in a list of lists

Question

I have the following structure:

list_of_tuples = [['portugal', np.nan, 'japan'], [np.nan, 'germany', 'canada'], ['UK', 'US', np.nan]]

I want to change the numpy nan values for MissingValue and turning back to a list of tuples:

final_structure = [('portugal', 'MissingValue', 'japan'), ('MissingValue', 'germany', 'canada'), ('UK', 'US', 'MissingValue')]

How can I replace the elements given a condition without iterating all over the elements? is there any single statement?

You can't; you have to examine each of the elements to determine which ones need changing. It's possible to write a one-liner, but it tsill must examine each value. — Prune, Jul 17 '18 at 18:04

Kefeng91 · Accepted Answer · 2018-07-17T17:49:05.717

There is no way to do it without iterating because tuples are immutable. However, you can use list comprehension to create a new list of tuples within a single line, following

final_structure = [ tuple([ j if isinstance(j, str) else "MissingValue" for j in l ]) for l in list_of_tuples ]

In your specific case, you can replace everything that is not a string (the np.nan) by "MissingValue" using the intrinsic function isinstance().

score 1 · Answer 2 · answered Jul 17 '18 at 17:48

This is one approach.

import numpy as np

list_of_tuples = [['portugal', np.nan, 'japan'], [np.nan, 'germany', 'canada'], ['UK', 'US', np.nan]]
list_of_tuples = map(lambda x: tuple(["MissingValue" if i is np.nan else i for i in x]), list_of_tuples)
print(list_of_tuples)

Output:

[('portugal', 'MissingValue', 'japan'), ('MissingValue', 'germany', 'canada'), ('UK', 'US', 'MissingValue')]

How can I change an element in a list of lists

2 Answers2