CODEWITHSUNDEEP
pythonpandasrename
TiiTcHY
TiiTcHY

Reputation: 69

Pandas Rename Columns

I am trying to rename my columns with the previous column name with a number on the end to make the columns unique. Is there a way to achieve this?

CurrentDF:

Reconnaissance      Unnamed: 1          Resource Development    Unnamed: 3      Initial Access          Unnamed: 5  
Active Scanning     Scanning IP Blocks  Acquire Infrastructure  Botnet          Drive-by Compromise         NaN

Desired:

Reconnaissance      Reconnaissance_1        Resource Development    Resource Development_1      Initial Access          Initial Access_1
Active Scanning     Scanning IP Blocks  Acquire Infrastructure  Botnet                      Drive-by Compromise         NaN

Upvotes: 3

Views: 543

Answers (3)

Yajat Malhotra
Yajat Malhotra

Reputation: 118

To rename all columns at once you can do the following:

df.columns = [col1, col2, col3]

Upvotes: 0

ALollz
ALollz

Reputation: 59579

You can create a Series from the columns (because Index objects have no ffill method, which is useful here). Then determine which columns start with Unnamed, mask them and use a cumcount to figure out what number to add onto the end (in the case of possibly multiple consecutive Unnamed: columns) and use ffill to get the previous column label that didn't start with 'Unnamed'. Assign the columns with this Series.

Sample Data

import pandas as pd
import numpy as np

df = pd.DataFrame(columns=['Reconnaissance', 'Unnamed: 1', 'Resource Development',    
                           'Unnamed: 3', 'Initial Access', 'Unnamed: 5'],
                  data=1, index=[0])

Code

s = pd.Series(df.columns)

s = s.mask(s.str.startswith('Unnamed:'))
s = (s.ffill() 
     + s.groupby(s.notnull().cumsum()).cumcount().astype(str).radd('_').replace('_0', ''))

df.columns = s

print(df)

   Reconnaissance  Reconnaissance_1  Resource Development  Resource Development_1  Initial Access  Initial Access_1
0               1                 1                     1                       1               1                 1

And here's another example to show how this behaves with less regularly spaced 'Unnamed:' columns.

df = pd.DataFrame(columns=['a', 'Unnamed: 1', 'Unnamed: 2', 'b', 'c', 'Unnamed: 3'],
                  data=[np.arange(6)], index=[0])

#### Same code as above

print(df)
   a  a_1  a_2  b  c  c_1
0  0    1    2  3  4    5

Upvotes: 3

not_speshal
not_speshal

Reputation: 23166

If it's every second column that needs renaming, you can use:

df = df.rename(columns = {df.columns[i]: f"{df.columns[i-1]}_1" for i in range(1, len(df.columns),2)})

Upvotes: 4

Related Questions