Home / Questions / 516644
Accepted
AlejandroR
Asked: 2022-07-24 02:06:57 +0800 CST 2022-07-24 02:06:57 +0800 CST

Remove text from a field to convert it to int in pandas dataframe

  • 772

I am trying to clean a column from a pandas dataframe. I have tried to convert an object of type object to int64:

df2['Pts'] = df2['Pts'].astype(int)

But it won't let me because there is a text string in these fields, they are all numeric but in some cases it appears like this: 55[a] or 55[b]

I have tried to remove those text strings that appear at the end of the numbers to be able to convert it, but it is not possible for me.

df2.replace(['[a]', '[b]'], ['', ''])

Could someone suggest me an idea so that those text strings do not appear inside my int values ​​to be able to operate with said data. Thank you very much in advance. All the best

python

1 Answers

  1. Best Answer

    You can do it in many different ways. I propose several:

    with regex

    import re

regex = re.compile(r'[^0-9]') # Eliminamos todo lo que no sean números
df['Pts'] = df['Pts'].replace(regex, '').astype(int)

    with extract

    Note: doing it this way you will only have the first string of numbers that appear consecutively

    df['Pts'] = df['Pts'].str.extract('(\d+)', expand=False).astype(int)

    With a lambda +isdigit

    df['Pts'] = df['Pts'].map(lambda x: ''.join([i for i in x if i.isdigit()])).astype(int)
    • 2