Python AttributeError：“系列”对象没有属性“ isdigit”

Question

I am trying replace number to '' blank if the row/column contains numbers. I tried the following it keeps complaing isdigit doesn't exist? I tried converting the column to string and doesn't help. Is there other operators that I could use for the pandas data frame?

data = ['123567','1547892','2547879','ABC','3D']
df1 = pd.DataFrame(data)
df1.columns = ['col1']

df1.col1 = str(df1.col1)
if len(df1.col1) < 8 and df1.col1.isdigit(): # errors
    df1.col1 == ''
    print(df1)

Looking for an output like this:

col1
0   
1   
2   
3   ABC
4   3D

Answer 1

To access string methods on a series, you need to do so via the .str attribute of Series :

df1.col1.str.isdigit()

See Series.str.isdigit() for the documentation.

You can use that as a boolean index and directly assign to the selected rows:

df1.col1[df1.col1.str.isdigit()] = ''

See Working with Text Data .

Do not use df1.col1.str.isdigit() in an if statement, because a boolean array is not True or False by itself, it is an array of boolean values and so would throw ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all() ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all() if used in a boolean context.

Demo:

>>> import pandas as pd
>>> data = ['123567','1547892','2547879','ABC','3D']
>>> df1 = pd.DataFrame(data)
>>> df1.columns = ['col1']
>>> df1
      col1
0   123567
1  1547892
2  2547879
3      ABC
4       3D
>>> df1.col1[df1.col1.str.isdigit()] = ''
>>> df1
  col1
0
1
2
3  ABC
4   3D