Change the Order of a Pandas DataFrame Columns in Python

Pandas may be a strong and well-known Python toolkit for data manipulation and analysis. When working with data, one common movement is to rearrange the columns of a DataFrame. It may be required for a variety of reasons, counting preparing data for introduction, guaranteeing interoperability with other frameworks, or just making data simpler to comprehend and examine. In this article, we'll see at various approaches for reordering columns in a Pandas DataFrame and show broad illustrations to help you master the task.

Changing the Order of Columns in a Pandas DataFrame

There are numerous methods for reordering columns in a Pandas DataFrame, each with advantages and instances where it is most appropriate. Here, we look at the theory and use of five common techniques:

To specify a new column order,
Use the '.reindex()' method.
Using the '.loc[]' indexer,
Using the '.insert()' method for reordering
Using the '.sort_index()' method.

1. Specifying a New Column Order Directly

This method entails generating a list of column names in the desired order and reassigning the DataFrame to this new order. It is uncomplicated and simple to apply.

The direct specification method involves explicitly defining the new column order. Pandas DataFrames offer column access by label. Thus, you can rearrange them by reassigning the DataFrame to the new order list.

Code:

 
import pandas as pd
# Sample DataFrame
data = {
    'A1': [1, 2, 3],
    'B1': [4, 5, 6],
    'C1': [7, 8, 9],
    'D1': [10, 11, 12]
}
df = pd.DataFrame(data)
# New column order
new_order = ['D', 'B', 'A', 'C']
# Reordering columns
df = df[new_order]
print(df)   

Output:

 
     D1  B1  A1  C1
0   10    4    1     7
1   11    5    2     8
2   12    6    3     9

2. Using the '.reindex()' Method

You can modify the order of columns using the .reindex() method by passing in a new list of columns. This technique provides more flexibility and alternatives for managing missing columns.

Pandas '.reindex()' method reassigns the Dataframe's columns based on the specified list. It can handle missing columns by either replacing them with NaN values or excluding them entirely, making it a versatile option.

Code:

 
# Reordering columns using reindex()
df = df.reindex(columns=['D1', 'B1', 'A1', 'C1'])
print(df)   

Output:

 
     D1  B1  A1  C1
0   10    4     1     7
1   11    5     2     8
2   12    6     3     9

3. Utilizing the '.loc[]' Indexer

The .loc[] indexer allows you to choose and reorder columns by providing the appropriate order within the indexer.

The .loc[] indexer in Pandas is a label-based indexer that selects rows and columns. By supplying a list of columns to .loc[], you can quickly reorder the DataFrame's columns.

Code:

 
# Reordering columns using .loc[]
df = df.loc[:, ['D1', 'B1', 'A1', 'C1']]
print(df)  

Output:

 
     D1  B1  A1  C1
0   10    4    1     7
1   11    5    2     8
2   12    6    3     9

4. Leveraging the .insert() Method for Reordering

The .insert() technique is useful for relocating a single column to a different position. This procedure requires deleting the column from its current place and inserting it at the appropriate index.

Pandas '.insert()' method allows you to insert a column at a precise spot. You can change the columns of a DataFrame one at a time by removing and reinserting them from their original positions.

Code:

 
# Sample DataFrame
data = {
    'A1': [1, 2, 3],
    'B1': [4, 5, 6],
    'C1': [7, 8, 9]
}
df = pd.DataFrame(data)
# Moving column 'C1' to the first position
column_to_move = df.pop('C1')
df.insert(0, 'C1', column_to_move)
print(df)   

Output:

 
    C1  A1  B1
0   7    1     4
1   8    2     5
2   9    3     6

5. Employing the .sort_index() Method

The .sort_index() method sorts columns according to their labels. It is useful if you want to sort columns alphabetically or numerically.

Pandas '.sort_index()' method sorts the DataFrame's columns by index label. It is especially handy for organising columns alphabetically or numerically without manually choosing the order.

Code:

 
# Reordering columns alphabetically
df = df.sort_index(axis=1)
print(df)   

Output:

 
    A1  B1  C1
0   1    4     7
1   2    5     8
2   3    6     9

Reordering columns in a Pandas DataFrame is a basic job that can improve data readability, interoperability, and presentation. You may efficiently manage and organise your data to satisfy varied needs by mastering techniques such as direct specification, '.reindex()', '.loc[]', '.insert()', and '.sort_index()'. Each methodology gives distinct preferences and adaptability, permitting you to choose the best alternative based on your individual needs. Understanding and adopting these procedures will permit you to handle data more productively, streamline information handling activities, and guarantee that your DataFrame is well-structured for examination and presentation. Whether you're preparing data for a report, integrating it with outside frameworks, or essentially making it simpler to work with, rearranging columns is an basic ability in your data manipulation tool kit.

Next TopicDifference between eq vs is vs in python

← prev next →