Input

   Column
0   2 mm
1   3 kg
2   4 m
3   name
4   2 mm
5   3 mph
6   full
7   left

Here I need to remove Units from the table. I tried with

df["Column"] = df["Column"].replace("\D", "", regex = True)

It gives me wrong output.

Expected Output:

    Column
0   2
1   3
2   4
3   name
4   2
5   3
6   full
7   left

CodePudding user response：

You can use

df["Column"] = df["Column"].str.replace(r'(\d)\s*[a-zA-Z] $', r'\1', regex=True)

See the regex demo. Regex details:

• (\d) - Group 1 (the \1 numbered backreference refers to this group value from the replacement pattern): any digit
• \s* - zero or more whitespaces
• [a-zA-Z] - one or more ASCII letters
• $ - end of string.

CodePudding user response：

You still can use your replace

s = df.Column.replace('[^0-9] ','',regex=True)
df.Column = df.Column.mask(s!='',s)
Out[27]: 
0          2
1          3
2          4
3       name
4          2
5          3
6       full
7       left
Name: Column, dtype: object

CodePudding user response：

You can use str.extract: if the row begins by a number ^\d , get it or | keep the entire row .*.

df['Column'] = df['Column'].str.extract(r'(^\d |.*)')
print(df)

# Output
  Column
0      2
1      3
2      4
3   name
4      2
5      3
6   full
7   left