In this tutorial, we are going to discuss how we can change the column order of a given pandas DataFrame object. During the data preprocessing stage, we might encounter a situation where the columns of the concerned pandas DataFrame are not in the desired order then we have to change the column order of the DataFrame.
Also read: Pandas DataFrame Indexing: Set the Index of a Pandas Dataframe
How to change the column order of a Pandas Dataframe?
Let’s get right into the different methods to change the column order of a dataframe in Pandas.
Method 1: Using the desired order columns list
This is one of the simplest methods to change the order of the columns of a pandas DataFrame object. In this method, we simply pass the Python list of columns of the DataFrame in the desired order to the DataFrame object. Let’s see how to code this method in Python.
# Method-1
# Import pandas Python module
import pandas as pd
# Create a pandas DataFrame
df = pd.DataFrame({'Roll': [111, 112, 113, 114, 115],
'Name': ['Sanjay', 'Aman', 'Ankit', 'Ravi', 'Komal'],
'Branch': ['ECE', 'ICE', 'IT', 'CSE', 'CHE'],
'CGPA': [8.15, 9.03, 7.85, 8.55, 9.45]})
print('Given pandas DataFrame:\n')
print(df)
# Change the order of the DataFrame
# Using the desired order columns list
df_1 = df[['Name', 'CGPA', 'Roll', 'Branch']]
print('\nPandas DataFrame with changed column order:\n')
print(df_1)
Output:
Given pandas DataFrame:
Roll Name Branch CGPA
0 111 Sanjay ECE 8.15
1 112 Aman ICE 9.03
2 113 Ankit IT 7.85
3 114 Ravi CSE 8.55
4 115 Komal CHE 9.45
Pandas DataFrame with changed column order:
Name CGPA Roll Branch
0 Sanjay 8.15 111 ECE
1 Aman 9.03 112 ICE
2 Ankit 7.85 113 IT
3 Ravi 8.55 114 CSE
4 Komal 9.45 115 CHE
Method 2: Using the loc method
In this method, we will make use of the loc method of the pandas DataFrame class. Using the loc method, we can reorder the columns of the pandas DataFrame object by providing a Python list of column names. Let’s write the Python code to implement this method.
# Method-2
# Import pandas Python module
import pandas as pd
# Create a pandas DataFrame
df = pd.DataFrame({'Name': ['Sanjay', 'Aman', 'Ankit', 'Ravi', 'Komal'],
'Roll': [111, 112, 113, 114, 115],
'Branch': ['ECE', 'ICE', 'IT', 'CSE', 'CHE'],
'CGPA': [8.15, 9.03, 7.85, 8.55, 9.45]})
print('Given pandas DataFrame:\n')
print(df)
# Change the order of the DataFrame
# Using the loc method of pandas DataFrame class
df_2 = df.loc[2:4, ['Roll', 'Name', 'CGPA', 'Branch']]
print('\nPandas DataFrame with changed column order:\n')
print(df_2)
Output:
Given pandas DataFrame:
Name Roll Branch CGPA
0 Sanjay 111 ECE 8.15
1 Aman 112 ICE 9.03
2 Ankit 113 IT 7.85
3 Ravi 114 CSE 8.55
4 Komal 115 CHE 9.45
Pandas DataFrame with changed column order:
Roll Name CGPA Branch
2 113 Ankit 7.85 IT
3 114 Ravi 8.55 CSE
4 115 Komal 9.45 CHE
Method 3: Using the iloc method
In this method, we will be using the iloc method of the pandas DataFrame class. Using the iloc method, we can reorder the columns of the pandas DataFrame object by providing a Python list of column indices (i.e. 0, 1, 2, 3, …) instead of the column names. Let’s see how to implement this method through Python code.
# Method-3
# Import pandas Python module
import pandas as pd
# Create a pandas DataFrame
df = pd.DataFrame({'CGPA': [8.15, 9.03, 7.85, 8.55, 9.45],
'Name': ['Sanjay', 'Aman', 'Ankit', 'Ravi', 'Komal'],
'Roll': [111, 112, 113, 114, 115],
'Branch': ['ECE', 'ICE', 'IT', 'CSE', 'CHE']})
print('Given pandas DataFrame:\n')
print(df)
# Change the order of the DataFrame
# Using the iloc method of pandas DataFrame class
df_3 = df.iloc[1:4, [1, 2, 0, 3]]
print('\nPandas DataFrame with changed column order:\n')
print(df_3)
Output:
Given pandas DataFrame:
CGPA Name Roll Branch
0 8.15 Sanjay 111 ECE
1 9.03 Aman 112 ICE
2 7.85 Ankit 113 IT
3 8.55 Ravi 114 CSE
4 9.45 Komal 115 CHE
Pandas DataFrame with changed column order:
Name Roll CGPA Branch
1 Aman 112 9.03 ICE
2 Ankit 113 7.85 IT
3 Ravi 114 8.55 CSE
NOTE: In the above two methods loc and iloc, we have an added advantage of selecting only a range of rows in the given pandas DataFrame object.
Method 4: Using the reindex() function
In this method, we will be using the reindex() function of the pandas DataFrame object. Using the reindex() function, we can rearrange the columns order of the pandas DataFrame object by passing a Python list of column names. Let’s implement this method through Python code.
# Method-4
# Import pandas Python module
import pandas as pd
# Create a pandas DataFrame
df = pd.DataFrame({'Branch': ['ECE', 'ICE', 'IT', 'CSE', 'CHE'],
'CGPA': [8.15, 9.03, 7.85, 8.55, 9.45],
'Name': ['Sanjay', 'Aman', 'Ankit', 'Ravi', 'Komal'],
'Roll': [111, 112, 113, 114, 115]})
print('Given pandas DataFrame:\n')
print(df)
# Change the order of the DataFrame
# Using the reindex() function of pandas DataFrame class
df_4 = df.reindex(columns = ['Roll', 'CGPA', 'Name', 'Branch'])
print('\nPandas DataFrame with changed column order:\n')
print(df_4)
Output:
Given pandas DataFrame:
Branch CGPA Name Roll
0 ECE 8.15 Sanjay 111
1 ICE 9.03 Aman 112
2 IT 7.85 Ankit 113
3 CSE 8.55 Ravi 114
4 CHE 9.45 Komal 115
Pandas DataFrame with changed column order:
Roll CGPA Name Branch
0 111 8.15 Sanjay ECE
1 112 9.03 Aman ICE
2 113 7.85 Ankit IT
3 114 8.55 Ravi CSE
4 115 9.45 Komal CHE
Summing-up
In this tutorial, we have learned how the four different ways to change the order of the columns of a pandas DataFrame object. Hope you have understood all the methods discussed above and are excited to use them on your own. Thanks for reading and stay tuned with us for more such amazing content on Python programming.