Adventures in Machine Learning

Swapping Columns in Pandas DataFrame: Quick and Easy Tutorial

Pandas DataFrame Column Swapping

Have you ever been working with a CSV or Excel file in Python and needed to swap the position of two columns in a DataFrame? The good news is that it is simple and quick to accomplish in just a few short steps using Python and the Pandas Library.

Here, we will look at how to swap two columns in a Pandas DataFrame and then provide a working example of the implementation of this task.

Swapping Two Columns in a DataFrame

Let’s start with the fundamental process of swapping columns in a Pandas DataFrame. To accomplish this task using Python, you will need to define a custom function that can handle swapping lists.

Here is a basic custom function in Python for swapping list items:


def swap_items_in_list(lst, item1, item2):

idx1, idx2 = lst.index(item1), lst.index(item2)

lst[idx1], lst[idx2] = lst[idx2], lst[idx1]

return lst


The next step is to correspond each column using Pandas DataFrame structure. To complete this, we will use Pandas DataFrame by using the .iloc function and pass a comma separated string with the indexes of the columns we want to switch.


def swap_columns(data_frame, col1, col2):

colnames = data_frame.columns.tolist()

i1, i2 = colnames.index(col1), colnames.index(col2)

colnames[i2], colnames[i1] = colnames[i1], colnames[i2]

return data_frame[colnames]


Let’s have a closer look at how the swap_columns function works. First, it receives the DataFrame and two columns that will be swapped.

Then, we store the list of column names existing in the DataFrame separately. After that, we reference the index of both columns we want to swap.

Therefore, the function replaces the name of column i1 with a name of i2 position and vice-versa. Finally, it ensures only the columns that are relevant to us are returned using the colnames argument.

Implementing the swap_columns Function

Now that we know how to swap two columns in a Pandas DataFrame, let’s move on to a more practical example to help apply this knowledge. As mentioned, we will use a sample DataFrame to help illustrate this task.

Here is an example DataFrame:


import pandas as pd

data = {‘team’: [‘atl’, ‘nyk’, ‘mia’, ‘chi’, ‘sac’],

‘rebounds’: [43, 44, 37, 42, 36],

‘points’: [102, 100, 89, 96, 82],

‘assists’: [25, 26, 20, 19, 21],

‘steals’: [10, 7, 6, 9, 11]}

df = pd.DataFrame(data)


With the above code, we have created a sample DataFrame object with team statistics that can be used to evaluate and analyze the team’s performance. It has the following fields and corresponding information: ‘team’, ‘rebounds’, ‘points’, ‘assists’ and ‘steals.’

Now, let’s define the swap_columns function using the code from earlier.


def swap_columns(data_frame, col1, col2):

colnames = data_frame.columns.tolist()

i1, i2 = colnames.index(col1), colnames.index(col2)

colnames[i1], colnames[i2] = colnames[i2], colnames[i1]

return data_frame[colnames]


In the example function above, the function takes in three arguments: the DataFrame, col1, and col2. The list of column names is gotten by calling the DataFrame’s columns method and converting it to a list using the tolist() method.

Afterward, the index of each column required is retrieved using the index() method, and then the swap is done using list swapping expressions. Swapping the ‘points’ and ‘rebounds’ Columns

Now that we have the necessary functions, it’s possible to carry out the task of swapping the ‘points’ and ‘rebounds’ columns.

Here is the Python code that swaps the required columns:


df = swap_columns(df, ‘points’, ‘rebounds’)



With the code above, the ‘points’ column has become the second column, whereas, previously, it was in the third position. You can see that weve managed to make a successful swap of the columns positions.


As stated earlier, swapping columns in a Pandas DataFrame is simple and quick, and this article breaks down how you can accomplish this task in a few easy steps. With the custom function defined in Python and Pandas DataFrame structure, you can switch columns that have various data types.

Hopefully, this information will help you make better data-driven decisions.

Additional Resources for Common Pandas Operations

Working with data presents a range of challenges that may require a broad range of tools and techniques. Fortunately, the Pandas library provides a fantastic solution to some of the common issues you might encounter when working with data in Python.

Pandas has a vast array of functionalities, and in this article, we will cover common Pandas operations and where to find detailed tutorials that cover them.

Overview of Common Pandas Operations

Pandas provides a comprehensive toolkit for data manipulation and analysis in Python. When it comes to data analysis in Python, there are a few common Pandas operations that you need to know:


Reading and Writing Data: Before performing any analysis with Pandas, you must first import data from a source and export the output. The data you access and manipulate could range from CSV, Excel, or database files to RESTful APIs. Pandas provides a series of functions that can read and write data from different sources such as pd.read_csv(), pd.read_excel(), pd.read_sql(), and pd.to_csv().

2. Selecting Data: In most workflows, you might only be interested in specific columns of a DataFrame.

Pandas provides a range of mechanisms to filter DataFrames, including indexing, loc, and iloc. These mechanisms can filter by rows, columns, and more advanced logics.

3. Filtering Data: After selecting your DataFrame, Pandas gives you options to filter data to a specific subset.

To filter data, use conditional expressions such as ==, !=, <, >, and others. 4.

Sorting and Grouping Data: Data can be sorted in Pandas using the sort_values() function. Sorting can be done based on specific columns, ascending, or descending orders.

Grouping is an essential function that groups data into an array of groupings based on specific columns. 5.

Transforming Data: It takes more than filtering, selecting, or aggregating a subset of data to get insights from it. Pandas provides mechanisms that can help to re-index, reshape, pivot, stack, and melt DataFrames to make them more suitable for analysis.

These are some fundamental transformation utilities that Pandas provides.

Detailed Tutorials on Performing Common Pandas Operations

Learning Pandas can be challenging and require time and dedication. Fortunately, there are a few online resources that can help you get started and strengthen your Pandas toolkit.

Here are some Pandas tutorials to help you get started:

1. Pandas Documentation: Pandas has one of the most complete documentations on data manipulation and analysis.

The Pandas documentation is an excellent resource for novice and advanced learners. It provides readers with informative explanations, examples, and even Jupyter notebooks that cover common Pandas operations.

2. Kaggle: Kaggle, the world’s largest data science community, provides a wealth of resources for data scientists and data enthusiasts.

The site has a wide range of Pandas tutorials that cover common use cases, including manipulating data, filtering data, and creating visualizations using Pandas. 3.

Real Python: The Real Python website provides a comprehensive and in-depth guide to using Pandas to analyze data in Python. The site’s tutorials break down how to perform data manipulation and analysis tasks using Pandas.

Its step-by-step approach makes it easy for learners to get started with Pandas analytics. 4.

DataCamp: DataCamp provides interactive data science courses that cover Pandas operations and other data manipulation techniques. Some of its popular Pandas courses that novice and advanced learners will find beneficial include “Pandas DataFrames” and “Manipulating DataFrames with pandas.”


Towards Data Science: Towards Data Science is a fantastic platform for data-related resources, including Pandas tutorials. It publishes high-quality, beginner-friendly, and in-depth guides on common Pandas operations.

Its community is also great for asking questions and seeking feedback on Pandas-related issues.


Pandas is the go-to toolkit for data manipulation and analysis in Python. In this article, we covered common Pandas operations, which are essential in most workflows.

We also highlighted some of the most beginner-friendly and comprehensive tutorials you can use to learn Pandas from various sources such as Pandas documentation, Kaggle, Real Python, DataCamp, and Towards Data Science. In conclusion, Pandas is an essential toolkit for data manipulation and analysis in Python.

Common Pandas operations include reading and writing data, selecting data, filtering data, sorting and grouping data, and transforming data. By mastering Pandas, you can simplify your workflow and gain valuable insights from your data.

With a range of online tutorials available on sites like Kaggle, DataCamp, Real Python, and Towards Data Science, it has never been easier to learn the ins and outs of Pandas. The takeaway is that investing your time in learning Pandas will significantly improve your data analysis skills, making you more productive and efficient in the workplace.

Popular Posts