How to swap values between two columns based on conditions in Python

August 28, 2024

I am trying to switch values between Range & Unit column from below dataframe based on condition if Unit contains - then replace Unit to Range & Range to Unit. To do that I am creating a unit_backup column so that I don’t loose the original Unit value.

1. dataframe

sample_df = pd.DataFrame({'Range':['34-67',12,'gm','45-90'],
              'Unit':['ml','35-50','10-100','mg']})
sample_df

    Range       Unit
0   34-67         ml
1   12         35-50
2   gm        10-100
3   45-90         mg

2. Function
Below is the code I have tried but I am getting error in this:

def range_unit_correction_fn(df):

    # creating backup of Unit column
    df['unit_backup'] = df['Unit']
    
    # condition check
    if df['unit_backup'].str.contains("-"):
        
        # if condition is True then replace `Unit` value with `Range` and `Range` with `unit_backup`
        df['Unit'] = df['Range']
        df['Range'] = df['unit_backup']
        
    else:
        # if condition False then keep the same value
        df['Range'] = df['Range']
    
    # drop the backup column
    df = df.drop(['unit_backup'],axis=1)
    
    return df

Applying above function on dataframe

sample_df = sample_df.apply(range_unit_correction_fn, axis=1)
sample_df

Error:

   1061 def apply_standard(self):
   1062     if self.engine == "python":
-> 1063         results, res_index = self.apply_series_generator()
...
----> 4     if df['unit_backup'].str.contains("-"):
      5         df['Unit'] = df['Range']
      6         df['Range'] = df['unit_backup']

AttributeError: 'str' object has no attribute 'str'

It seems like some silly mistake but I am not sure where am I going wrong.

Appreciate any sort of help here.

>Solution :

When you access df['unit_backup'], you get a scalar string value, not a pandas Series, so calling .str on it raises an error.

To fix it you can check the condition directly on the string value in a row-wise approach:

def range_unit_correction_fn(df):
    # creating backup of Unit column
    df['unit_backup'] = df['Unit']

    # condition check
    if '-' in df['unit_backup']:
        # if condition is True then replace `Unit` value with `Range` and `Range` with `unit_backup`
        df['Unit'] = df['Range']
        df['Range'] = df['unit_backup']

    # drop the backup column
    df = df.drop(['unit_backup'])

    return df