3

I have a text file containing a column of numbers:

10              
20               
40              
13               
24                
35               
44

I convert this to a pandas column, and try to convert it to string. But for some reason I can't seem to do so.

import pandas as pd

df=pd.read_csv('file.txt')
df.columns=['column1']
df['column1']=df['column1'].astype(str)
print(df['column1'])

This gives:

0    10
1    20
2    40
3    13
4    24
5    35
6    44
Name: column1, dtype: object

The dtype is still object isntead of string. Don't quite know why this is the case, since astype should convert it to string.

samman
  • 559
  • 10
  • 29
  • Does this answer your question? [Convert Columns to String in Pandas](https://stackoverflow.com/questions/22005911/convert-columns-to-string-in-pandas) – Red Jul 09 '20 at 01:44
  • No, since I already knew how to convert it to a string, I was just confused as to the printout and nomenclature. – samman Jul 09 '20 at 14:10

1 Answers1

2

That is how pandas define the column type , there is not string type column, it belong to object

df.column1.apply(type)
0    <class 'str'>
1    <class 'str'>
2    <class 'str'>
3    <class 'str'>
4    <class 'str'>
5    <class 'str'>
Name: column1, dtype: object

DataFrame dose not str.replace

You should do

df.replace({'...':'...'})

Or

df['column1']=df['column1'].str.replace()
BENY
  • 317,841
  • 20
  • 164
  • 234
  • 1
    The problem is when I try to do ```df['column1']=df.str.replace()```, I get an error ```AttributeError: 'DataFrame' object has no attribute 'str'``` implying the dataframe is still not a string. This is why i was trying to convert it to a string – samman Jul 09 '20 at 01:21
  • 1
    @samman that is for series not dataframe , I have update my answer – BENY Jul 09 '20 at 01:23
  • So do I even need to change the types to str? To try your method? Also what's the difference between df.replace and df[].str.replace – samman Jul 09 '20 at 01:26