How to change an element from string to float in a nested list

Question

I want to use list comprehension to change an element within a nested list to float.

lines = [[1, '74.37000326528938', 'Psychologist'], [2, '67.49686206937491', 'Psychologist'], [3, '74.92356434760966', 'Psychologist']]
>>> lines = [[line[0]] + [float(x) for x in line[1]] + [line[2]] for line in lines]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: could not convert string to float: .

I also don't understand why this would work for a different example:

mylist = [['1','2','3'],['5','6','7']]
datl3 = [[int(x) for x in line[0]] + line[1:] for line in mylist]
datl4 = [[line[0]] + [float(x) for x in line[1]] + [line[2]] for line in datl3]
>>> datl4
[[1, 2.0, '3'], [5, 6.0, '7']]

I think it has to do with decimal points, because the code won't work for this example:

mylist = [['1','2.555','3'],['5','6.777','7']]
datl3 = [[int(x) for x in line[0]] + line[1:] for line in mylist]
datl4 = [[line[0]] + [float(x) for x in line[1]] + [line[2]] for line in datl3]
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: could not convert string to float: .

Why are you building a new list? Shouldn't for l in lines: l[1] = float(l[1]) work? — cs95
– cs95, Commented Sep 30, 2018 at 0:30
No, since l is a reference to an inner list, it ends up modifying the original lines. — cs95
– cs95, Commented Sep 30, 2018 at 0:33

Sheldore · Accepted Answer · 2018-09-30 01:05:12Z

1

You were close but the problem was that you were using a for loop on line[1] which is a single string. Hence, you were looping through each element/character of your float number. When you reach the decimal . you get the ValueError: could not convert string to float: .. Your second example works because there you just have a single digit strings 2 and 6 without any decimal.

If you stick to your code, you just need to replace the index line[1] by line[1:2] as follows where you slice

lines = [[line[0]] + [float(x) for x in line[1:2]] + [line[2]] for line in lines]

However, a more concise way is following where you avoid an unnecessary intermediary for loop by directly accessing the second element of the sublists. You don't need to create three sublists and then add them using +

lines = [[line[0], float(line[1]), line[2]] for line in lines]

Output

[[1, 74.37000326528938, 'Psychologist'],
 [2, 67.49686206937491, 'Psychologist'],
 [3, 74.92356434760966, 'Psychologist']]

edited Sep 30, 2018 at 1:05

answered Sep 30, 2018 at 0:34

Sheldore

39.2k9 gold badges63 silver badges76 bronze badges

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

3 Comments

Code-Apprentice Over a year ago

for x in line[1:2] is a complicated way of getting line[1].

Sheldore Over a year ago

@Code-Apprentice: I agree but my aim here was to tell the OP what is going wrong in his/her approach. Reading all three answers, I feel we all are telling the same thing about the decimal

Code-Apprentice Over a year ago

You could add to your question by showing how to write the final solution in a more clear and concise way.

Code-Apprentice · Accepted Answer · 2018-09-30 00:40:26Z

1

The problem is that this line is trying to do too many things at once:

datl4 = [[line[0]] + [float(x) for x in line[1]] + [line[2]] for line in datl3]

The problem occurs in the inner list comprehension:

[float(x) for x in line[1]]

But what is line[1]? It comes from the iteration for line in datl3. So let's see what values it has:

>>> [line[1] for line in datl3]
['2.555', '6.777']

So when you do [float(x) for x in line[1]] you are iterating over the characters of each numerical string..."2" then "." then "5" then "5" then "5". But "." is not a valid float.

I think what you want to do instead is just convert line[1] to a float:

lines = [[line[0], float(line[1]), line[2]] for line in lines]

answered Sep 30, 2018 at 0:40

Code-Apprentice

84k26 gold badges162 silver badges289 bronze badges

Comments

vash_the_stampede · Accepted Answer · 2018-10-01 00:54:21Z

0

It can be done with list comprehension just need to set it up properly, check the str for a decimal, if so make that item an int

l = [[float(j) if '.' in str(j) else j for j in i] for i in l]

[[1, 74.37000326528938, 'Psychologist'], [2, 67.49686206937491, 'Psychologist'], [3, 74.92356434760966, 'Psychologist']]

answered Oct 1, 2018 at 0:54

vash_the_stampede

4,5861 gold badge11 silver badges20 bronze badges

Collectives™ on Stack Overflow

How to change an element from string to float in a nested list

3 Answers 3

3 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

3 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related