将列表列表中除一个之外的所有字符串更改为浮点数

Question

I have a list of lists:

LofL = [["string", "number", "number", "number"], ["string", "number", "number", "number"], ...]

and im trying to turn the numbers within the list of lists to floats instead of strings. Each list of list is not equal in length. So far I have:

for sublist in LofL:
sublist[1:] = float(sublist[1:])

but I'm getting the error float argument must be a string or real number not a list . This works for single numbers if I do sublist[1] = float(sublist[1]) but I'm unsure of how to include all numbers in each list of lists without the [1:] indexing.

Answer 1

for sublist in LofL:
    for i in range(1, len(sublist)):
        sublist[i]=float(sublist[i])

Explanation: Your initial code sublist[1:]=float(sublist[1:] will not work as sublist[1:] is a list and you cannot turn list to a float.

So the for i in range(1, len(sublist) will iterate through each element in your sublist then with each iteration we will convert sublist[i]=float(sublist[i] into float. Because you specified that the first item of sublist does not need to change hence we are starting our loop from 1

Answer 2

This is very similar to the other answer here. I'm just including it because it seems like the reassignment of a slice of the list is what you were going for in your original code:

LofL = [["string", '1', '2', '3'], ["string", '4', '5', '6', '7']]

for sublist in LofL:
    sublist[1:] = map(float, sublist[1:])

print(LofL)
# [['string', 1.0, 2.0, 3.0], ['string', 4.0, 5.0, 6.0, 7.0]] 

Answer 3

The excellent answer from @Mark solves your particular problem. However, there is an implicit assumption that the first element of each sub-list should be ignored and that subsequent elements can be converted to float.

Here's a simple double loop that will try to convert all elements. This would allow for the 'string' component to be anywhere in the list.

LofL = [["string", "42", "42", "1"], ["string", "2.5", "100", "20"], [5, 'string']]

for e in LofL:
    for i, v in enumerate(e):
        try:
            e[i] = float(v)
        except ValueError:
            pass

print(LofL)

Output:

[['string', 42.0, 42.0, 1.0], ['string', 2.5, 100.0, 20.0], [5.0, 'string']]