Pandas apply method | String recognised as a float

Question

I have to retreat a dataframe's column where all observations contain the same type of value, a string as follows :

'Total: 1'
'Total: 1'
'Total: 2'
etc.

Then I try to run the apply method like this to the column evolLevel which contains those strings :

df['evolLevel'].apply(lambda x: x[-1])

However it returns me the following error although this is still a string variable :

'float' object is not subscriptable

I've also tried to specify axis=1 but it doesn't change anything.

I do no get any error when I run the code you are running. It works just fine (even `axis=1` is not needed). — moys, Mar 09 '21 at 12:53
That's weird, but if finally works if y do df['evolLevel'] = df['evolLevel'].apply(lambda x: str(x)[-1]) — GaëtanLF, Mar 09 '21 at 12:57
@MayankPorwal no it won't, if the column contains a `float` object it will fail in exactly the same way — juanpa.arrivillaga, Mar 09 '21 at 13:15
@GaëtanLF that means your column has a `float` object in it. — juanpa.arrivillaga, Mar 09 '21 at 13:16

score 1 · Accepted Answer · answered Mar 09 '21 at 12:57

1

To whom it may be interesting, it finally worked when I did :

df['evolLevel'] = df['evolLevel'].apply(lambda x: str(x)[-1])

answered Mar 09 '21 at 12:57

GaëtanLF

1 Answers1